Examining the India, Brazil and
South African (IBSA) Triangular
Trading Relationship
by
Ron Sandrey and Hans G. Jensen
National Agricultural Marketing Council document – 2007
This paper is prepared for the National Agricultural Marketing Council (NAMC) by Trade Law
Centre for Southern Africa (Tralac).
The financial support from AusAID for the GTAP modeling is acknowledged.
Please note that this paper can also be accessed from the Tralac website as Tralac Working
Papers
1
Examining the India, Brazil and South African Triangular Trading Relationship
Ron Sandrey and Hans Grinsted Jensen
Summary and general conclusions from the analysis
Following a comprehensive examination of the most recent merchandise trade flows
between the relevant countries, this paper uses a computer model to look at the possible
economic results from removing all merchandise1 tariff barriers between the three partners of
India, Brazil and South Africa/SACU (IBSA). We have ignored the political complication of
Brazil also belonging to a Free Trade Agreement (FTA) and we have not modelled the
estimation and removal of non-tariff barriers, services trade or some of the more
sophisticated but speculative gains from technological change or other dynamic effects.
Recent research, some of which is cited in the paper, has highlighted that most of these
more elaborate assumptions are misleading to policy makers.
The IBSA agreement is potentially good for all major parties with similar welfare gains of
between one to one and a half billion dollars at 2015, but with this translating into larger
gains for South Africa when measured as a percentage of real GDP since South Africa has a
smaller economic base to work from. The gains to South Africa are spread across the
contributing factors of allocative efficiency, labour’s contribution, capital and the terms of
trade gains from both (a) better relative prices between exports and imports and (b) more
efficient use of capital. The biggest loser in dollar terms is the EU, with all other
countries/regions except Nigeria losing. Unfortunately these losers include both Botswana
and the Rest of SACU or the model aggregation of Lesotho, Namibia and Swaziland
combined, although these losses are not high and may be misleading given that intra-SACU
trade and therefore any changes in this trade will not be picked up in the model’s database in
view of the poor quality of this trade data.
Another feature of the analysis is that we have used as our base for the simulation a trade
picture that includes all the known global updates, and this includes simulating the effects of
the Trade, Development and Cooperation Agreement (TDCA) with the EU in such a way that
it enables us to isolate these effects from the base. Results from this TDCA simulation
suggest that a full and comprehensive IBSA FTA is of greater value (in fact, about double the
1
We have used the interchangeable terms of ‘merchandise’, ‘goods’ and ‘products’ either together or
separately in this paper. They generally refer to the actual physical items that are traded. The
celebrated definition from The Economist is that these are items that hurt when dropped on your foot,
and the terms are used to preclude the trade in services.
2
welfare gains) to South Africa than the partial TDCA as it now stands. This is mainly because
(a) South Africa faces manufacturing tariffs that are modest, thus the preferences are not that
significant, and, more importantly, (b) South Africa gains little preference into the highly
protected European agricultural market from the TDCA. Conversely, for IBSA, South Africa is
deemed to have gained comprehensive access into the relatively highly protected Indian
market, thus gaining a considerable advantage over global competitors in both agricultural
and non-agricultural goods.
In the first section of the chapter we examine the current trade flows between the IBSA
partners and hypothesise that the interesting results for South Africa may concentrate upon
the sugar trade in agriculture and the motor vehicle trade in the non-agricultural sectors.
Neither of these proved to be significant for South Africa. Sugar production actually declines
in South Africa despite gaining better access into India, as this access is taken up by Brazil
rather than the presumably less efficient South African production. The case is similar for
motor vehicles, where South African production declines by 1.6 percent in the face of more
efficient production, and consequently imports from Brazil in particular and, to a lesser
extent, from India.
Given the extent to which China, with its dramatically increasing exports over the last few
years, has displaced South African domestic production of clothing, it should come as no
surprise that India, although currently not a major source of South African imports but a
country with enormous production capacity perhaps second only to China, should compete
strongly in South Africa if tariffs were to be eliminated. Clothing production declines by
11.0 percent, and this is a massive decline for an individual sector2.
However, the major finding from this GTAP exercise, and one not anticipated from the trade
data, is the massive gains to South Africa from attractive access into India from a zero rather
than 15 percent duty on gold. This is a happy juxtaposition of the world’s leading gold
producer with a large jewellery exporter that enables both partners to prosper since India’s
costs are reduced. Indeed, it is this sector that is driving a considerable portion of the welfare
gains to both South Africa and India, and the policy implication is very clear: reducing the
Indian tariffs on gold is a win-win situation and must become a priority for negotiators.
Another major finding from this GTAP work is related to the employment closures, where the
trade-offs between holding wages constant and increasing the supply of unskilled workers at
2
Sandrey (2006c) in discussing the trade and economic implications of the South African restrictions
regime on the imports from China provides extensive background on the clothing imports into South
Africa.
3
one extreme and, conversely, holding the supply of workers constant and increasing wages
are clearly demonstrated. The somewhat intermediate position of allowing the unskilled
labour increase to be a function of the unemployment rate in each country has been adopted
as the standard closure for the model, with these two extremes and another closure that
holds wage rate increases for unskilled workers to be in line with the inflation rate. There is
no doubt that holding wages down and reflecting as much of the change in increasing
unskilled workers entering the labour force is the best policy for South Africa. This increases
welfare and lessens inflationary pressures quite substantially.
We have provided two alternative scenarios to judge the full and complete removal of
merchandise tariffs. These are (a) a 50 percent tariff reduction rather than the full
100 percent, and (b) a full 100 percent IBSA simulation post-Doha. Results for (a) show that
this offers 43 percent of the gains for South Africa and a lesser 30 percent for Brazil, but is
much better for India which maintains 62 percent of its full gains as the relative prices move
around, in consequence of which the trade outcomes are not a linear 50 percent. Results
from (b) show that since the Doha results are modest, their diminution of the original IBSA
100 percent results are similarly modest for South Africa.
Results for the agricultural sector are modest. Initial agricultural products have been a very
minor part of South Africa’s exports into India’s heavily protected market, while agricultural
imports from India are concentrated in the duty-free imports of rice. Brazil has become a
major global player in agricultural exports, and sends large quantities of soya bean products
and poultry meats, pork and beef to South Africa. Following the FTA South Africa increases
exports to India by $182 million and Brazil by an insignificant $7 million. Overall some
$144 million of the increase is trade diversion from previous destinations and leaves a global
increase of only $44 million overall. Increases in vegetable oils and fats ($70 million) and
wool ($63 million) to India take place, while there are global reductions in exports of (a)
vegetables, fruit and nuts and (b) other food products. For imports, there is a similar but
slightly larger overall increase of $92 million, driven mostly by increased imports from Brazil
of $76 million (other crops, other meats, and vegetable oils and fats).
The implication for the BLNS countries (Botswana, Lesotho, Namibia and Swaziland) are
disquieting, as they see declines in their welfare. This mostly comes from terms of trade
losses as the better access for South African non-agricultural goods into India consequently
increases the relative prices for SACU imports from South Africa. Exports of sugar products
(we presume from Swaziland) to India increase, but this is mostly at the expense of reduced
exports to the EU overall. Exports from Botswana reduce marginally in the manufacturing
4
sector as their costs increase (also marginally). Imports from India increase, but almost all of
this is a substitution away from the traditionally-based South African source. There are very
low and insignificant changes in the trade flows with Brazil. In agriculture, there are no (or
almost no) changes to trade flows other than the sugar exports to India.
Finally, we undertake some alternative scenarios around the unskilled labour market closure
assumptions in the primary model. We expand from the standard assumption that
employment is fixed and the adjustment is through the wage rate, to use the closure whereby
unskilled labour supply is a function of the unemployment rates in each country, and the
adjustment therefore varies between changes in employment and the wage rate depending
upon that initial unemployment rate. We also simulate a scenario where the closure has the
real wage fixed and all adjustments must come through the number of unskilled persons
employed. Here the results are striking: employment is up by 2.83 percent, welfare more
than doubles from the primary model results to $3,006 million and the Consumer Price Index
(CPI) is increasing by a lesser 0.43 percent. This dramatic result clearly highlights that if
South Africa is serious about increasing both welfare and employment in the economy, the
more policies move towards creating jobs rather than rewarding those actually in
employment is a superior option for policy makers.
Introduction and background
In recent months, the question of a closer trading relationship between India, Brazil and
South Africa (the three IBSA partners) has received much media attention. The objective of
this chapter is to focus upon the current merchandise trade and place this in perspective for
a more cooperative approach. The analysis will extend to undertaking an advanced computer
general equilibrium modelling study to assess what the gains may be from a trilateral FTA
between the partners. The chapter will note here at the beginning that such an FTA is
conceptual only, as both Brazil and South Africa are members of their own FTAs – Mercosur
and SACU respectively – and a country cannot be a member of more than one customs
union. Nonetheless, an FTA sets the outer boundaries for an enhanced trading regime, and
in recognition of South Africa’s membership of the Southern African Customs Union (SACU)
we will consider the implications of the IBSA agreement for the fellow SACU members of
Botswana, Lesotho, Namibia and Swaziland (BLNS). All data in this first section of the paper
is sourced from the commercially available World Trade Atlas3 (Dr John Brasher), which is in
turn sourced from the country’s official sources.
3
World Trade Atlas at www.gtis.com.
5
The analysis will concentrate upon using the respective partners’ import data, as this is
generally more reliable than export data. Two points must be made here, however. The first
is that imports are usually assessed using CIF (the value of the goods plus the costs of
freight and insurance transporting them from the export dock to be unloaded at the import
border) while exports are assessed as FOB (free on board, or the actual value of the goods
at the export dock). This means that we are in effect inflating the import value as compared
with the export value by including freight costs, and in the case of some bulk products, this is
significant. However, it is important to note that South Africa is one of the few countries in the
world that both reports on and assesses duty for imports without adding the costs of
shipment and insurance. The second point is that export and import values seldom if ever
agree, and we will introduce a reconciliation exercise to explore some reasons for this, with
the CIF versus FOB values as just one reason. All data is expressed in US dollars.
Overall all three partners share important and somewhat equal trading relationships.
Firstly, South Africa’s major exports to the world are concentrated in minerals and related
products. The major global exports by the general HS 2 Chapter4 has been precious metals
and stones, etc. (platinum, gold and diamonds), followed by iron and steel products, mineral
fuels and motor vehicles, and these four products made up 55.6 percent of the total exports
during 2005. The main export to India during 2005 was the one-off aircraft, followed by
inorganic chemicals, iron and steel products and again precious metals and stones. These
top four accounted for 65.5 percent of the total South African exports to India, and in addition
we would note that Indian import data shows that precious stones, metals and minerals, etc.
may be underreported in the export data as South Africa does not generally disclose its
export destinations for gold. Our analysis suggests that these exports face an average tariff
of some 15.88 percent using Indian import data. The main exports to Brazil were iron and
steel, organic chemicals, general machinery and mineral fuels, with these four comprising
63.3 percent of the total. Again, tralac analysis suggests that these products faced an
average duty of 8.95 percent during 2005, and that the proposed SACU/Mercosur FTA would
make little difference to this rate.
4
Where HS is the Harmonised System of merchandise trade classification that operates in a
sequentially more detailed level from internationally harmonised (hence the name) HS 2 to 4 and 6
levels, and often down to even HS 10 for individual countries. For example, the HS6 classification
scheme contains about 5000 product groups, and this and often more detailed levels are used for
specifying tariff schedules.
6
Secondly, India’s main exports to the world are precious metals and stones, mineral fuels,
clothing and organic chemicals, with these four making up a lesser 36.3 percent of the total.
There is a direct linkage between India’s exports of precious jewellery and South Africa’s
base metal exports, as India has become the main processing and trading centre globally for
these often South Africa-sourced raw materials. Exports to South Africa were concentrated in
mineral fuels, vehicles, cereals and iron and steel products, with these exports making up
52.9 percent of the total. Indian exports to Brazil were mineral fuels (49.6% of the total),
organic chemicals, pharmaceuticals and miscellaneous chemical products, with the top four
comprising 71.3 percent of the total.
Finally, Brazilian global exports are even more diversified than India’s, with the top four of
vehicles, machinery, iron and steel and ores, etc., comprising 32.0 percent of the total.
Exports to South Africa are concentrated in vehicles and parts (31.8%), machinery, meats
(poultry) and sugars. In fifth place is soya bean oil and in sixth place is tobacco, reaffirming
the importance of Brazil for South African agricultural imports. Exports to India feature
sugars, soya bean products, aircraft and beverages, with these making up 63.2 percent of
the total. In agriculture, Mercosur in general provides around one-quarter of the global
imports, and Brazil is highly competitive in the crucial beef, pig meat and poultry sectors with
all three (along with sugar cane) receiving virtually no government supports.
It is noticeable that iron and steel products and vehicles and their associated parts feature in
the top four global exports from both Brazil and South Africa, and these are ranked sixth and
eighth respectively from India. There appears to be a degree of intra-industry trade among
the three partners in these products as well, leading to the general conclusion that the gains
from cooperation here may not necessarily be in ‘new’ trade but in more sophisticated
linkages in existing products. Brazil, as a major agricultural exporter, will remain a valuable
source of both food for direct consumption in South Africa and imports such as soya beans to
provide the feedstuffs for the livestock sector.
The automobile sectors
Of particular interest to South Africa has to be the automobile sector, where the new
international ‘buzz word’ acronym is ‘BRIC’ for Brazil, Russia, India and China. Analysts
consider that the biggest breakthrough in global growth will come from these BRIC countries.
They will shortly account for more than 40 percent of forecast global light vehicle assembly
increases and represent around half of the industry’s forecast global capacity expansion.
Consequently, nearly all major global automakers are pursuing a BRIC strategy in some form
7
as they attempt to gain competitive advantage by linking a presence into these emerging
markets5. South Africa could be poised to gain an advantage over several competitors should
some form of preferential and cooperative trading agreement be formulated between IBSA.
The Brazilian sector in particular is very competitive, despite currently facing excess
capacity, high rates of taxation and interest and a weakening consumer demand. However,
its technology in the flex-fuel engines which run on gasoline, ethanol or any blend of the two
is likely to create a huge demand for this technology should global oil prices stay high.
Similarly, India’s annual automobile sector growth has been the highest in the world in recent
years, with this accentuated by the shift in automobile production from the US and Western
Europe to Asia, with India and China taking the lead. Consequently, India has become the
largest manufacturer of tractors in the world, the second largest manufacturer of threewheelers and two-wheelers, third largest manufacturer of commercial vehicles, and fifth
largest manufacturer of cars. International companies are scrambling to establish themselves
in this market in particular, and that presence includes establishing design and research
centres in addition to India’s more traditional lower cost production.
The agricultural sectors
Since the early 1990s India has undergone considerable economic policy reform, although
agricultural reform has lagged behind other sectors. The general pattern of agricultural
protection (as measured by the Producer Support Estimate (PSE)) is for support to rise when
world prices are low and decline when they are high, making analysis very complex.
Depending upon the method used to calculate these PSEs, they were near their highest in
2002 at 11.0 to 19.2 percent overall, while in 1996 they were negative (i.e. taxing the sector)
at similar rates.
Agricultural exports have been a very minor part of South Africa’s exports to India: during
2005 they were some 1.8 percent of the South African exports to India, a similar figure to
the1.9 percent of 2003 but below the 2004 figure of 3.5 percent when a larger export of sugar
was made. Indeed, some 87 percent of the agricultural exports during 2005 were either
sugar (50.2%) or wool (36.8%), while fresh fruits and cotton provided another 4.7 percent
and 2.0 percent respectively. This pattern has changed little over the last ten years. Sugar
appears to have potential for increased imports from South Africa under a less distorted
regime, notwithstanding the fact that India is the world’s second largest producer of sugar
5
[Online]. Available:
http://www.pwc.com/Extweb/ncpressrelease.nsf/docid/7C7BE1A291BB6BD5852572040082059C.
8
after Brazil with around 15 percent of the global production. Between 2001 and 2003 India
was a large exporter of sugar, but in 1999, 2004 and 2005 it has been an importer. Brazil is
certainly more internationally competitive in sugar than India and even probably South Africa.
Meanwhile, India’s sugar regime is highly regulated, with an import duty of 100 percent plus
a possible countervailing duty of another 850 rupees per ton, the sugar levy obligations, the
sugar release quota system and other domestic regulations. Thus, there seems to be some
scope for cooperation in this agricultural market in particular.
Brazil has virtually replaced Australia as the international champion for agricultural free trade
globally. Its potential is enormous, with good land exceeded only by China, the US and
Australia, and a policy regime that has both contributed to and benefited from the radical
economic reforms in Brazil over the last 15 years. Support to the sector is now minimal (PSE
of around 3%, a figure even lower than South Africa’s and not much above the radical
agricultural reformers of New Zealand and to a lesser extent Australia). Overall production
has similarly increased following the economic reforms, with technological change a key
driver in this expansion, an expansion that has recently lead to concerns about the
environmental consequences of agriculture moving into new lands. Brazilian agricultural
exports have increased to the extent that they are now around 30 percent of total exports,
and they moved dramatically away from the traditional tropical products of coffee and orange
juice to being major global suppliers of soya beans, sugar and the grain-soya bean feed for
chickens, cattle and pigs.
The structure of the chapter is as follows. The first section will examine the current bilateral
relationships sequentially, starting with (a) South Africa and India before examining (b) South
Africa and Brazil and then (c) Brazil and India. This section will be shortened, and the reader
is referred to Sandrey and Jensen (2007 6 ) for a fuller discussion on these trading
relationships, including more details on assessed tariffs, growth rates and market share to
put this trade in perspective before examining the future with a ‘trade chilling’ analysis of
potential trade products based on the current trade patterns for the respective partners. The
second section of the chapter will use the Global Trade Analysis Project (GTAP) international
computer model to simulate the impacts of an FTA between the partners under different
scenarios.
Section 1: The trading relationships
6
Sandrey, R. & Jensen, H. 2007. Examining the India, Brazil and South African (IBSA) Triangular
Trading Relationship. tralac Working Paper 1. [Online]. Available: www. tralac.org.
9
The big picture
South Africa, India and Brazil are emerging globally in that Brazil and India have both
become crucial players in the world trading arena while South Africa is the important ‘bridge’
between the developed world and developing Africa. In terms of economic power, the World
Bank ranks Brazil 10th, India 12th and South Africa 27th by traditional GDP, but all are
higher by the more useful Purchasing Power Parity (PPP or what your money will buy) index.
As a measure of import openness, Brazil collected import duties equivalent to 8.4 percent of
total merchandise imports, while the similar figures for India and South Africa were 14.1 and
2.7 percent respectively. Thus, South Africa could be regarded as being very open on
average (with motor vehicles and clothing distorting that average) while India is relatively
highly protected. Brazil is a major agricultural exporter with one-third of its total exports
classified as agriculture (compared to India at 11.3% and South Africa at 7.9%), while
agricultural imports range between 6.2 to 7.0 percent in all three cases.7
All three partners are also part of what may be described as ‘second rung’ trading nations,
and all rank relatively closely together, with world ratings between 24th for both Brazil in
world exports and India in world imports, and 37th for South Africa in world exports.
Examining the World Trade Atlas data shows global exports from Brazil, South Africa and
India of $118 billion, $52 billion and $100 billion respectively during 2005, and similarly,
imports of $74 billion, $55 billion and $138 billion for the same countries. An analysis of the
WTO Annual Report shows very similar but not always identical data; for South African
imports, for example, the WTO reports $67 billion rather than $55 billion. We would
hypothesise that the difference is that the World Trade Atlas does not report intra-SACU
trade for South Africa. Table 1 shows the general trading relationships between the three
countries, along with the bilateral rankings based upon individual destinations/sources (i.e.
with the EU countries ranked on their own). The data confirms that the respective partners
are (not in a derogatory manner) ‘second rung’ but still important trading partners ranging
from 11th for South African imports from Brazil to 36th for Brazilian imports from
South Africa.
Table 1: the bilateral trading relationships, US$ million
Country
2005 rank
1999
2001
2003
2004
2005
South African exports to
India
14
386
379
394
574
1,169
Brazil
30
155
275
177
242
317
7
WTO at www.wto.org.
10
South Africa imports from
India
13
247
243
419
712
1,102
Brazil
11
225
383
714
1,001
1,306
South Africa
18
290
321
448
883
1,420
Brazil
25
128
227
383
542
976
South Africa
15
1,777
1,395
1,933
1,767
2,651
Brazil
28
306
267
306
662
879
South Africa
21
237
424
733
1,036
1,369
India
23
314
285
553
652
1,137
South Africa
36
172
286
202
268
342
India
17
170
543
486
556
1,203
Indian exports to
Indian imports from
Brazilian exports to
Brazilian imports from
Source: World Trade Atlas
Table 1 hints at some differences in the trade data between the respective partners and
suggests why import rather than export data is used to assess likely tariff rates, for example.
This is not to say that one country’s data is better than another, but rather that factors such
as transshipment, adding the costs of freight and insurance to imports, and a general
comment that imports are likely to be more scrutinised than exports all point towards using
import data, especially when using the World Trade Atlas data for all three countries. Table 2
extracts the relevant data from Table 1 to enable this comparison to be made. For the South
African export/Indian import data there are significant differences, and these are explored
later in the chapter and in more depth in Sandrey (2006a8). Most other pairings are in the
‘close enough’ category. We would again note that, unlike most other countries, South Africa
values imports and assesses the duty on what is called FOB basis rather than the more
normal CIF. This can make a difference by increasing the value of imports by perhaps an
average of around ten percent or more, but in many cases much higher depending upon
relative efficiencies at the port and the costs of shipping costs relative to actual costs of bulk
products like iron ore. Having noted these factors, since we are assessing the relationships
from a South African perspective, we will concentrate upon the South African data in general
except when assessing tariffs at the partner border.
8
Sandrey, R. 2006. South African Merchandise Trade with India. tralac Working Paper No 10.
Stellenbosch: US Printers.
11
Table 2: The trade flow data from respective partners 2005, $m
India exports
Brazil imports
976
India exports
1,203
RSA imports
1,420
Brazil exports
1,102
RSA imports
1,369
1,306
Brazil exports
India imports
1,137
RSA exports
879
India imports
1,169
RSA exports
2,651
Brazil imports
317
342
Source: World Trade Atlas
The South African/Indian relationship
This section presents a summary and the main points reported in earlier tralac research by
Sandrey (2006a), to which the reader is referred for more details.
(a) South African exports/Indian imports
During 2005, South Africa exported merchandise worth $1,169 million to India, almost
doubling the 2004 figure of $574 million. This represented some 2.3 percent of South Africa’s
global exports during 2005, a figure that similarly doubled from the previous year (and
basically the ten-year average). This is shown in Figure 1, with the big increase in 2005
obvious. Not shown is that India ranked as South Africa’s 14th most important export
destination, just behind Italy and Zimbabwe but ahead of France, Mozambique and Korea.
12
Figure 1: South African exports to India, $ million and % share
1,400
2.5%
1,200
% sha re
2.0%
$m
1,000
800
1.5%
600
1.0%
$m
% share
400
0.5%
200
0
0.0%
1996
1998
2000
2002
2005
Source: World Trade Atlas data
The increase in 2005 was mostly aircraft ($248 million), virtually a new export line as only in
2003 ($4.8 million) had aircraft previously featured at all. Other main exports at the
HS 2-chapter level were inorganic chemicals ($179.7 million), iron and steel ($170 million),
precious stones ($168 million) and fuels (also $168 million). Other than inorganic chemicals,
all of these exports more than doubled from their 2004 values. Of particular importance are
the exports of HS 28, inorganic chemicals. These exports are shown as predominantly HS
2809, diphosphorus pentaoxide, phosphoric acid, etc., and they comprise 91.2 percent of the
total global exports of these chemicals from South Africa.
During 2005, the mirror of imports into India from South Africa totalled some
US$2,651 million, a figure that increased by 50.1 percent from 2004. In ‘real’ terms, this
represented 1.9 percent of Indian global imports, a figure that has shown a downward trend
over the last seven years. By commodities, the main imports were gold ($1,846 million),
coal/oil/gas (($171 million), and pentaoxide chemicals ($149 million). The average duty on
these imports, calculated using the Indian Tariff Schedule, was 15.88 percent. These imports
into India are highly concentrated, with the top four lines at the HS 2-chapter level accounting
for 90.8 percent of the total trade, and in some of the main imports at the more detailed level,
these imports held a significant share in the Indian market. South Africa is doing relatively
well into India, with some 45.0 percent of the imports (the ‘Stars’) gaining market share in
Indian import sectors that are themselves growing.
Data reconciliation shows that, as expected with South African imports into India valued at
over double the reported exports from South Africa to India, there is little coherence in
13
reconciling the trade flows to India. Some of this may be accounted for in the transport costs
of the bulkier exports such as iron and steel products, given that imports are valued at CIF
and exports at FOB, while another large difference is that there are no reported exports of
gold from South Africa but this is the major import into India. In addition, the largest export
from South Africa (aircraft) is not reported in Indian import data. If we ignore this gold and
aircraft trade then the overall reconciliation is remarkably accurate.
(b)
Indian imports into South Africa
Indian imports into South Africa by December 2005 were $1,102 million or 2.0 percent of
the total global South African imports, with the latter figure increasing dramatically over the
last three years (Figure 2). These Indian imports are not as highly concentrated as South
African imports into India, with the top five HS chapters of motor vehicles, fuels, cereals,
organic chemicals and pharmaceuticals making up a lesser 47 percent of the imports.
Although 49.9 percent of the trade enters duty-free, the average duty that would have been
assessed on these imports in the absence of any rebates was 11.01 percent. These duties
are mostly paid by the 30 percent of the imports that enter where the duties are assessed at
20 percent or more, a grouping that contains motor vehicles and their parts, and textiles,
clothing and footwear.
1,200
2.5%
1,000
2.0%
800
1.5%
600
1.0%
400
% share
$m
Figure 2: Indian imports into South Africa, $ million and % share
India $m
% Share
0.5%
200
0
0.0%
1996
1998
2000
2002
2005
Source: WTA
To set the scene for later analysis, India’s performance in the South African market relative
to Korea and Brazil is shown in Fig 3. India and Brazil were both slow starters, with a one
percent share in 1996, but since 2000 Brazil has been consistently ahead of India. Korea has
been consistently above India over the period.
14
Figure 3: Relative performance of India into South Africa, % market share
% share
3.0%
2.5%
2.0%
Korea
1.5%
Brazil
1.0%
India
0.5%
0.0%
1996
1998
2000
2002
2005
Source: World Trade Atlas
The South African/Brazilian trading relationship
(a) South African exports/Brazilian imports
During 2005, South Africa exported merchandise worth $317 million to Brazil, a figure that
has fluctuated over the last ten years but is very similar to the earlier 1997 figure of
$307 million. This represented some 0.6 percent of South Africa’s global exports during
2005, a figure that similarly fluctuated but generally trended downwards over the period. This
is shown in Figure 4, with US$ million values on the bars using the left-hand scale and the
percentage share of South African global exports destined for Brazil shown with the lines set
against the right-hand scale. The graph is interesting in that from 1996 through to and
including 2002 the relationship between the value of the exports to Brazil and their relative
share of South African global exports stayed almost exactly the same. From 2003 this close
relationship was broken as the values grew faster than the percentage, highlighting just how
fast South African global exports were increasing.
15
350
1.2%
300
1.0%
$m
250
0.8%
200
0.6%
150
0.4%
100
% share
Figure 4: South African exports to Brazil, $ million and export share
$m
% tot
0.2%
50
0
0.0%
1996
1998
2000
2002
2005
Source: World Trade Atlas
Table 3 shows the main products that are exported to Brazil by the HS 4 level general
category, with the list ranked by total exports over the last ten years. Coal/oil/gas has been
the most consistent product, followed by the top 2005 export of ferroalloys, a product whose
exports have consistently increased in the most recent two years, and the relatively
consistent insecticides. Others such as ethyl alcohol and phosphoric acid were important in
the early years, while still others such as aircraft and nickel plates have been more of a oneoff or less consistent trade. An analysis of the export data reveals that there are only two
instances where exports to Brazil at the HS 4 level are (a) above $10 million and (b)
represent at least 10 percent of South African exports. These are (i) HS 2934 nucleic acids,
$13.21 million and 45.8 percent of the exports and (ii) HS 5402 synthetic yarns,
$11.44 million and 13.2 percent of the exports. The tariffs that are faced by these products at
the Brazilian border are examined later using Brazilian import data.
Table 3: South African exports to Brazil, $ million
Description/Year
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
Total
272
307
197
155
200
275
171
177
242
317
Coal/oil/gas
35.7
49.7
51.5
37.1
29.8
34.8
28.4
22.5
26.9
29.1
9.5
9.1
6.7
5.2
8.9
12.7
12.5
25.3
50.3
50.6
Insecticides, etc
13.2
20.2
12.5
12.9
23.6
33.9
21.3
19.9
11.3
7.9
ethyl alcohol
73.7
60.0
0.0
1.2
5.7
0.2
0.0
0.0
0.0
0.0
aluminium plates
0.0
0.0
0.4
3.8
12.5
16.7
8.7
15.3
13.7
17.3
phosphoric acid
22.8
27.5
17.2
5.0
3.3
5.0
0.7
0.0
0.0
0.0
0.0
0.0
0.3
1.4
0.0
66.1
4.7
0.0
0.0
0.4
ferroalloys
aircraft
16
Description/Year
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
Total
272
307
197
155
200
275
171
177
242
317
synthetic yarn
3.3
9.7
5.3
4.6
7.2
6.5
6.0
6.4
8.4
11.4
nickel plates
0.0
0.0
8.1
14.2
17.7
6.9
6.2
0.0
0.0
1.5
ketones
4.0
1.8
1.0
2.9
3.7
2.3
3.2
6.0
7.7
10.5
engine parts
0.0
0.1
0.3
0.1
1.1
3.0
3.0
13.5
10.4
10.6
stainless steel
5.3
6.5
4.2
2.2
4.3
4.9
5.9
2.1
1.6
0.5
hydrocarbons
0.4
1.0
0.8
0.9
2.2
3.9
2.9
3.5
6.2
12.4
nucleic acid
2.9
3.0
0.0
0.0
0.0
0.0
0.0
0.0
5.5
13.2
Source: World Trade Atlas
As mentioned above, reconciliation of trade data is often a problem, so we have provided an
alternative Brazilian view with its imports from South Africa over recent years. The annual
totals by product are surprisingly consistent, with one notable exception. That exception is
the reported imports of platinum, Brazil’s second major import and a consistent import over
the years, which is not matched by reported exports from South African data. An examination
of Brazilian data shows that between 30 to 40 percent of the platinum imports over the last
three years have been from South Africa with almost all the remainder from the EU. We
would hypothesise that this is transshipment through Europe to Brazil.
Table 4: Brazilian imports from South Africa, $ million
Description
1997
1999
2001
2003
2004
2005
total
Totals
351
172
286
202
268
342
2,318
Coal/oil/gas
73.2
39.8
28.5
27.0
31.3
32.5
379.4
Platinum
21.0
23.9
73.8
19.1
29.0
33.8
274.5
ferroalloys
9.7
4.5
11.8
20.9
38.2
56.9
170.5
insecticides
17.4
12.3
24.7
11.8
6.3
7.2
119.5
ethyl alcohol
71.3
2.7
2.9
0.0
0.0
0.0
85.7
aluminium
0.0
2.3
15.7
14.9
13.3
13.1
83.5
nucleic acids
8.9
0.0
11.7
7.8
13.7
13.2
81.8
phosphoric acid
32.3
8.0
7.5
1.0
0.0
0.0
73.1
synthetic yarn
10.3
5.6
6.8
6.2
9.5
9.2
69.9
nickel
1.0
16.5
14.6
0.0
0.0
2.8
68.3
engine parts
0.0
0.0
2.4
12.5
10.0
8.0
36.2
ketones
1.5
2.0
2.3
4.0
7.7
9.5
36.1
Source: World Trade Atlas
17
Tariffs at the Brazilian border
We start this analysis by assessing the potential impact of the SACU/Mercosur FTA upon the
tariffs using the limited data that is available. This analysis shows that during 2005 only
16.6 percent of the reported South Africa imports would be afforded tariff preferences, with
almost all of this a complete elimination of the relevant but very selected tariff line. The
products concerned are almost exclusively coal/oil/gas (duty-free anyway), hydrocarbons
and the nucleic acids. Thus, the preferences under the SACU/Mercosur FTA seem very
restricted at best9.
The more complete analysis of South African tariffs faced at the Brazilian border presented
below is based upon Brazilian import data and the Brazilian tariffs as supplied from the
MacMaps database - with all analyses at the HS 6 line level. We note that the MacMaps tariff
data is for 2001, but it is unlikely that much would have changed since then. The data shows
that over three years (2003, 2004 and 2005) and using the 2001 tariff, duties on South
African imports into Brazil would have been 8.07 percent, 7.81 percent and 8.95 percent
respectively on average. Coal/oil/gas, the main import by value, is duty-free and furthermore
it is the only import line of the top 59 lines with import values of at least $1,000,000 that is
duty-free. The details for the main imports are shown below in Table 5 by HS 2-chapter.
Duties range from 0.2 percent for coal oil gas to 16.2 percent for man-made fibres. Not
shown is that some of the HS 2-chapters face higher tariffs; beverages at 23.2 percent is the
highest, while toys, etc., other fibres, shoes, clothing, leather and clocks, etc. all face rates
between 20.0 and 21.5 percent.
Table 5: Main South African imports into Brazil, $ million and duty faced (%)
Imports $ million
HS 2
Description
2003
2004
2005 Duty
2005
Av %
72
Iron/steel
25.5
43.5
82.1
9.5%
29
Organic chemicals
20.6
33.3
42.8
9.9%
27
Coal/oil/gas
38.0
41.3
36.0
0.2%
84
Machinery
15.2
17.0
34.8
16.0%
71
Precious stones
19.1
28.0
33.9
3.5%
76
Aluminium
14.9
13.9
16.7
11.3%
9
We agree with Stern and Flatters (2005) who are particularly scathing on this agreement. They write
that Mercosur conceded improved access into South Africa for 46 of their currently traded product
lines, and that South Africa’s imports of these products from Brazil and Argentina in 2004 ‘was a
miserly R26 million’. Overall, they considered that it would be hard to construct a less meaningful
agreement.
18
Imports $ million
HS 2
Description
26
Ores, etc
38
2003
2004
2005 Duty
2005
Av %
4.8
14.2
16.7
3.5%
Chemical products
14.0
11.3
10.7
10.3%
81
Base metal product
5.2
7.1
9.9
6.0%
54
Man-made fibres
6.2
9.5
9.3
16.2%
Source: World Trade Atlas data and MacMaps tariffs
(b) Imports from Brazil into South Africa
Brazil was South Africa’s 11th main individual country source of imports during 2005, one
place ahead of Australia and two ahead of India. These imports have climbed from
$225 million in 1996 to $1,306 million in 2005, or in ‘real’ terms from 1.0 percent to
2.4 percent. The data is shown in Figure 5, where the increase from 2000 can be seen.
Figure 5: South African imports from Brazil, $ million and % market share
1400
2.5%
1200
2.0%
800
1.5%
600
1.0%
% share
$m
1000
$m
%
400
0.5%
200
0
0.0%
1996
1998
2000
2002
2004 2005
Source: World Trade Atlas
Table 6 shows that the increases in imports have been driven by the top three products of
motor vehicles (HS 9801, with other vehicles also included in the table in the HS 87 codes),
poultry meat and soya bean oil.
During 2005, ignoring rebates, an average duty of
14.86 percent would have been assessed on these imports, with the total (theoretical) duty
bill being $194.03 million. This duty is heavily influenced by the high rates on vehicles,
although poultry meat, sugar confectionery and tobacco products all face high duties.
19
Table 6: South African imports from Brazil, $ million and assessed duty %
Descrip/yr
Total
vehicles
poultry
soya bean oil
iron products
buses, etc.
pork
confectionary
vehicle parts
tractors
motor cars
tobacco
vehicles
computers
bulldozer, etc.
engine parts
Subtotal
1996
1998
2000
2002
2003
2004
2005
Duty%
262
230
292
467
714
1,001
1,306
14.9
39.35
45.36
57.12
89.38
210.24
264.65
353.62
26.0
1.48
3.12
8.61
18.47
37.04
88.75
111.04
17.9
0.57
3.10
0.47
15.37
39.16
47.67
67.68
10.0
0.00
3.14
13.49
17.15
20.65
23.16
34.30
0.0
3.41
1.63
2.41
14.74
20.42
24.72
31.35
22.5
0.00
0.00
0.76
0.04
7.74
17.96
30.94
4.5
0.04
0.66
3.02
3.37
9.38
19.55
27.50
25.0
4.73
4.33
5.99
5.42
9.02
13.15
25.38
18.1
9.09
14.40
5.01
10.11
17.04
24.79
20.14
6.5
0.18
0.01
0.07
0.06
0.01
0.18
19.96
31.1
9.42
10.24
5.99
10.05
15.74
40.38
17.03
15.0
0.02
0.00
0.00
2.41
2.10
8.41
15.84
18.4
0.01
0.02
0.01
0.01
5.53
21.50
15.79
0.0
2.45
0.42
8.73
10.93
12.09
13.70
15.74
0.7
7.98
5.50
10.13
11.58
11.56
12.62
15.30
0.0
79
92
122
209
418
621
802
Source: World Trade Atlas data and tralac calculations
A feature of the South African agricultural imports in recent years has been the dramatic
rise in imports from South America, and the percent share of these agricultural imports from
both Argentina and Brazil are shown below in Figure 6. The imports from Brazil are mostly
the soya bean products and poultry meats, although pork and beef in the meat products and
sugar confectionery have made dramatic increases over the last three years10. Agricultural
imports from Argentina are focused on soya bean oilcakes, wheat and soya bean oil.
10
On the last three, we would not deny that there has been double-digit growth in the South African
imports of sugar products and pork over the period 1996 to 2005, while the imports of beef actually
declined by 1.4 percent over the same time. This suggests that the beef is merely import
displacement.
20
Figure 6: Brazil and Argentina, % of South African agricultural imports
20%
18%
16%
14%
12%
10%
8%
6%
4%
2%
0%
Arg
Brazil
1994
1996
1998
2000
2002
2004
Source: South African Revenue Service data
SACU and Mercosur have recently negotiated a preferential trade agreement, and this
marginally reduces some of the duties on imports into South Africa/SACU. Calculations by
tralac show that overall this agreement would have reduced the duties assessed at the South
African border by a modest $6.88 million to an average of 14.33 percent. Detailed analysis
reveals that these reductions are focused (in order of duty reductions) on refrigerators,
plastic sheets, generators, dental products, pork, coffee, food preparations not specified,
nuts, ceramic tiles and engine parts. These products in total account for 64 percent of the
tariff reductions, and the scope and magnitude of these reductions suggests that the
SACU/Mercosur trade agreement cannot be heralded as a major trade policy breakthrough.
As in the case of exports to Brazil, we again examined imports to assess where imports from
Brazil at the HS 4 level are (a) above $10 million and (b) represent at least ten percent of
South African imports. The first two products on the list are poultry products and soya bean
oils, where there are both significant imports and Brazil has a dominant market share, while
the third, a particular line of iron products, is sourced exclusively from Brazil. Thus, Brazil is
again by this measure more important to South Africa as an import source than an export
destination.
21
Table 7: Brazilian share of South African imports, % share and % million
Brazilian share RSA imports
product/yr
2003
2004
Brazilian imports $m
2005
2003
2004
2005
poultry
49.8%
78.0%
75.4%
37.0%
88.8%
111.0%
soya bean oil
60.0%
47.0%
61.5%
39.2%
47.7%
67.7%
iron products
92.5%
99.4%
99.8%
20.6%
23.2%
34.3%
buses
66.5%
64.6%
33.9%
20.4%
24.7%
31.3%
pork
42.9%
53.9%
65.4%
7.7%
18.0%
30.9%
confectionary
42.5%
40.8%
56.2%
9.4%
19.6%
27.5%
tractors
8.5%
10.7%
11.3%
17.0%
24.8%
20.1%
tobacco
20.3%
34.6%
21.9%
15.7%
40.4%
17.0%
tiles
28.0%
24.6%
20.6%
10.7%
14.5%
15.2%
generators
11.0%
10.8%
12.3%
8.8%
10.8%
14.4%
frozen beef
40.8%
52.0%
45.9%
3.5%
10.4%
13.6%
Source: World Trade Atlas
The Indian/Brazilian trading relationship
Table 8 replicates the relevant parts of Table 1 to show the aggregate data for this
relationship. As this ‘triangular’ trading relationship is of lesser importance to South Africa we
report only the general picture here. It is replicated here to remind the reader of the relative
importance to both Brazil and India of their bilateral trade flows and thus place the potential
FTA in perspective for these two countries.
Table 8: The bilateral trading relationships, US $ million
Country
2005 rank
1999
2001
2003
2004
2005
Southbound – India to Brazil
India export
25
128
227
383
542
976
Brazil import
17
170
543
486
556
1,203
Northbound – Brazil to India
India import
28
306
267
306
662
879
Brazil export
23
314
285
553
652
1,137
Source: World Trade Atlas
Indian imports from Brazil
Table 9 shows the main Indian imports from Brazil at the HS 4 level. Three agricultural
products are at the top: sugar, soya bean oil and ethyl alcohol. Sugar imports have varied,
22
while soya bean oil has been consistent and ethyl alcohol a new import. Many of the next
major imports are ores and chemical products, although there are some manufacturing
products such as pumps and refrigerators as well.
Table 9: Indian imports from Brazil, $ million
Description
1999
2000
2001
2002
2003
2004
2005
Total
305.6
181.2
266.8
332.5
306.4
661.9
879.1
sugar
96.2
35.2
0.0
0.8
0.0
127.9
186.5
soya bean oil
97.2
23.0
104.0
145.9
121.1
151.4
179.1
ethyl alcohol
0.0
0.0
0.0
2.1
5.1
99.6
147.3
iron ores
2.5
0.0
5.0
9.4
14.9
37.3
27.9
pumps
0.5
0.8
0.5
0.9
1.1
6.8
23.7
nickle
0.9
6.4
8.7
21.4
8.9
9.2
18.7
precious stones
5.6
6.5
4.1
8.6
8.0
11.8
17.0
asbestos
8.6
7.4
5.0
5.8
5.3
6.8
14.8
hydrocarbons
7.5
0.0
1.9
0.0
6.7
6.3
14.2
rubber
1.3
3.4
1.5
5.2
9.3
13.5
12.7
nitrile compounds
0.8
0.0
10.1
11.5
8.3
5.5
10.7
hydrocarbons
1.1
0.0
0.1
0.5
3.2
0.2
10.5
special
0.4
1.4
10.0
1.7
0.0
1.9
10.0
Source: World Trade Atlas
Brazilian imports from India
The main products imported into Brazil from India are shown in table 10 below, again at the
HS 4 level. These products are also concentrated in the fuels and chemical products, with oil
being the dominant import.
Table 10: Brazilian imports from India, $ million
Description
1997
1999
2001
2003
2004
2005
Total
216
170
543
486
556
1203
oil
0.0
0.0
249.8
212.7
162.9
618.0
medicants
0.5
1.6
20.5
27.1
40.5
48.9
nucleic acids
0.5
2.4
19.2
11.1
19.4
46.3
synthetic yarn
3.8
2.2
3.0
16.7
28.0
38.9
11.1
10.2
11.7
22.9
30.7
37.5
8.8
16.5
16.4
17.8
24.0
27.6
antibiotics
heterocyclic
23
Description
1997
1999
2001
2003
2004
2005
polyether
0.0
0.0
0.5
2.2
2.3
18.4
coloring agents
5.7
6.5
7.5
12.8
15.7
17.8
plastic sheets
2.0
4.1
6.1
3.8
8.4
15.1
coal/oil/gas/coke
0.0
0.0
0.0
0.0
8.2
14.7
carboxyyamide
0.9
1.2
33.9
5.3
5.6
13.2
amine comps
6.5
5.0
7.3
8.2
8.1
11.1
cyclic alcohols
2.7
1.3
2.2
4.5
10.3
9.4
Source: World Trade Atlas
Data reconciliation
In this chapter we have so far only reported on the import data for each partner, and this
short section will comment upon the relationships between this import data and the export
data as supplied by the other partner. Depending upon how the value of transport and
insurance associated with getting goods to market is assessed, we would expect that imports
would be 10 percent or more above the relevant export figure. This is the case for
southbound (Indian exports to Brazil – Brazilian imports from India) trade but not northbound
trade (the converse – Brazilian exports to India and Indian imports from Brazil).
For the southbound trade (Indian exports, Brazilian imports), the value of imports at $1,202.7
million is some 27 percent above the Indian export figure of $976 million as shown above.
Brazilian imports of HS 29, organic chemicals, are vastly underreported as exports from India
to Brazil, and this seems to be a big factor in influencing the overall total.
For northbound trade (Brazilian exports, Indian imports) the Indian import figure is only
77 percent of the Brazilian export figure. Examination of the data reveals that two of the five
main exports from Brazil, aircraft and fuel, represent some $210 million of exports but are not
reported as imports into India. This makes a big difference in rectifying the overall data. We
would note that aircraft are not an infrequent cause of data problems, as sometimes
confusion may exist between one country classifying the transaction as an exports and
another as a lease, with the latter not included in merchandise trade data11. Of the other main
exports, sugar is underreported into India (or over-reported from Brazil?) while soya bean oils
are also inclined that way, but the ethyl alcohol trade seems correct. We have not examined
quantity data.
11
We understand that this is probably what happened. In 2005, SAA leased an almost brand-new
Airbus to an Indian airline.
24
Overall conclusions
All three partners share important and somewhat equal trading relationships.
Firstly, South Africa’s major exports to the world are concentrated in minerals and related
products. The major global exports by the general HS 2-chapter has been precious stones,
etc. (platinum, gold and diamonds), followed by iron and steel products, mineral fuels and
motor vehicles, and these four products made up 55.6 percent of the total exports during
2005. The main export to India during 2005 was the one-off aircraft, followed by inorganic
chemicals, iron and steel products and, again, precious stones. These top four accounted for
65.5 percent of the total South African exports to India, and in addition we would note that
Indian import data shows the precious stones, metals and minerals, etc., may be
underreported in the export data. Our analysis suggests that these exports face an average
tariff of some 15.88 percent using Indian import data. The main exports to Brazil were iron
and steel, organic chemicals, general machinery and mineral fuels, with these four
comprising 63.3 percent of the total. Again, tralac analysis suggests these products faced an
average duty of 8.95 percent during 2005, and that the proposed SACU/Mercosur FTA would
make little difference to this rate.
Secondly, India’s main exports to the world are precious metals and stones, mineral fuels,
clothing and organic chemicals, with these four making up a lesser 36.3 percent of the total.
There is a direct linkage between India’s exports of precious jewellery and South Africa’s
base metal exports, as India has become the main processing and trading centre globally for
these often South African sourced raw materials. Exports to South Africa were concentrated
in mineral fuels, vehicles, cereals and iron and steel products, with these exports making up
52.9 percent of the total. Indian exports to Brazil were mineral fuels (49.6% of the total),
organic chemicals, pharmaceuticals and miscellaneous chemical products, with the top four
comprising 71.3 percent of the total.
Finally, Brazilian global exports are even more diversified than India’s, with the top four of
vehicles, machinery, iron and steel and ores, etc., comprising 32.0 percent of the total.
Exports to South Africa are concentrated in vehicles and parts (31.8%), machinery, meats
(poultry) and sugars. In fifth place is soya bean oils and in sixth place is tobacco, reaffirming
the importance of Brazil for South African agricultural imports. Exports to India feature
sugars, soya bean products, aircraft and beverages, with these making up 63.2 percent of
the total.
25
It is noticeable that iron and steel products and vehicles and their associated parts feature in
the top four global exports from both Brazil and South Africa, and these are ranked sixth and
eighth respectively from India. There appears to be a degree of intra-industry trade among
the three partners in these products as well, leading to the general conclusion that the gains
from cooperation here may not necessarily be in ‘new’ trade but are more sophisticated
linkages in existing products.
For the agricultural sector, Brazil is a major agricultural exporter and will remain a valuable
source of both food such as poultry, pork and beef for direct consumption in South Africa,
and imports such as soya beans to provide the feedstuffs for the livestock sector. As the
world’s major cane sugar producer, Brazil will also set the benchmark for South Africa’s
aspirations in the global market place, and as an unsubsidised agricultural producer this
statement becomes more critical and can be extended to the agricultural sector in general for
South Africa (witness the increases in red meat exports, for example). India remains a
relatively higher protected agricultural producer and suffers from internationally competitive
pressures as a result. It consequently may become a more important market for South Africa
should meaningful preferences be negotiated, but is unlikely to expand much as an
agricultural imports source. There are, however, intriguing potential dynamics for all three
countries in the sugar market as the world’s main producer (Brazil) is potentially linking with
both India and South Africa, two other major producers, and in the case of India, a periodic
importer in a trilateral where the protectionist markets of India and South Africa may be
opened.
The BLNS countries
The current trading relationship
These trading relationships are shown in Table 11, using the Indian and Brazilian trade data
for 2005. We have found in other trade data analyses that (a) the BLNS data is not generally
as up-to-date as the World Trade Atlas data, and (b) there is a degree of consistency using
BLNS country data but not really a good one-to-one mapping. We have therefore used the
Indian and Brazilian data only, but recognise, of course, that imports in particular are entering
the BLNS countries via South Africa.
26
Table 11: India/Brazil trade with BLNS, $ million and main HS 4 lines 2005
Indian Imports from
$ million
Main lines
Indian exports to
$ million
Main lines
B
0.4
radar equipment, machines
11.7
medicine, rubber, aluminium
L
0.0
none
16.1
fabric, tractors, pumps
N
3.2
scrap iron, engines, balloons
14.1
medicine, motorcycles
S
3.7
gold, electrical parts, acids
Brazilian imports from
B
0
L
4.0
medicine, chemical,
Brazilian exports to
na
2.3
tyres, stoves, ceramics
0.05
T-shirts
1.3
fabric (only exp)
N
0.02
car parts
12.9
furniture, cement, bulldozer
S
0.31
converters machinery
0.5
pump, tubes, confectionery
Source: World Trade Atlas
Direct imports into Brazil from the BLNS are very low, while the comparable imports into
India are also very low to modest ($3.7 million being the highest). Indian exports to BLNS are
higher, with medicines and vehicles being prominent, while Brazilian exports to Namibia are
the only significant trade in that direction. Trade data from the BLNS is somewhat consistent
with these figures, although Botswana’s exports to India are reported to be some
$12.1 million for 2005 (mostly gold).
In general, these direct reported flows are low, but, of course, there is no way of knowing
what the value of imports in particular into BLNS that are transiting through South Africa may
be. These low reported values will almost by definition ensure that an IBSA trade agreement
will have limited direct gains for the BLNS.
Section 2: The simulation
Introduction
The objective is to simulate the impact of possible multilateral market access reforms
resulting from an FTA between SACU, India and Brazil. The analysis is undertaken using a
variant of the GTAP model as discussed in Chapter 2, a discussion that includes the data
used, the assumptions and limitations of the model. In particular, note that in the process of
updating the Indian MFN tariffs it was observed that South Africa exports of coal/oil/gas was
classified as import under HS6 code 270112 Bituminous coal/oil/gas in the initial 2001
MacMaps HS6 digit tariff data, while in 2001 to 2005 South Africa was not reporting exports
27
under this code. Therefore the initial MacMaps databases MFN tariff on South African
coal/oil/gas exports to India was adjusted to reflect South Africa’s reporting of HS6 digit
codes on coal/oil/gas exports.
The FTA primary scenario entails the result from the removal of trade barriers between
India, Brazil and the member countries of SACU as measured in the year 2015 in a world
shaped by the baseline scenario. This implies that all ad valorem tariffs and ad valorem
equivalent of specific tariffs between participating countries India, Brazil and SACU (IBSA)
are abolished. Differences between the so-called baseline scenario and this so-called
primary scenario are therefore the results of implementation of the IBSA FTA. Note that we
are not modelling reductions in either services or any non-tariff barriers. In addition, it is
always possible to do an almost endless number of ‘what if' scenarios. For this study we
have limited these to two basic assumptions/possible outcomes. One is that there will be a
successful conclusion to the Doha Development Round of the WTO, while the other is that
the IBSA negotiators will fail to negotiate a fully comprehensive FTA, and we have proxied
this possibility by assuming the outcome will be a 50 percent cut in bilateral tariff rates. In
reality, it is more likely that sensitive sectors for either party will be isolated from any final
agreement, but we have opted for this blanket 50 percent option. To complete the picture we
have combined the latter two scenarios into a combined one of a successful WTO Doha
round increasing multilateral market access and the bilateral IBSA FTA.
Operationally, these are represented as follows:
•
Firstly, scenario S1 (the primary scenario) is run to simulate the effects of the
comprehensive FTA in a post-UR environment where all other known international
commitments are fully implemented;
•
Scenario S2 implements the partial FTA with a 50 percent reduction in applied ad
valorem equivalents (AVE) tariffs between the IBSA countries;
•
Scenario S3 is then run. This is an implementation of a stylised outcome of a possible
DDA round of the WTO using the assumptions outlined above;
•
Next, Scenario S4 extends S3 by also including the post-Doha FTA between IBSA
countries with a 100 per cent removal of AVE tariffs between the FTA countries.
In order to find the isolated effects of creating the FTA in a post DDA round, scenario S3 is
subtracted from scenario S4, quantifying the effect of the FTA. Therefore the results shown
later for the FTA (in an environment where the DDA has been completed) are technically the
results of scenarios S4 – S3.
28
Before we introduce the results of the simulations, we will present Table 12 with the tariffs
faced by South Africa into both India and Brazil and by both India and Brazil into South Africa
to give a perspective of the relative protection levels by GTAP sectors. This information is
reproduced in Annex Tables A1 and A2 along with the changes in trade flows as a result of
IBSA. Note that these tariff rates are as used within the GTAP model, and reflect the
composition differences in individual products within a particular GTAP sector from a
particular source at 2001, and therefore may not be directly comparable with the actual tariffs
on recent imports into South Africa as discussed earlier in the chapter. Note also that we
have not displayed the tariff levels for paddy rice, cane and beet sugar, and raw milk, as
these products are effectively non-traded. The table highlights that there are several sectors
in all countries that have relatively high initial levels of tariff protection.
Table 12: Initial tariff levels as at the baseline by GTAP sector (%)
Tariffs faced, South Africa into
Primary
Wheat
India
Brazil
Tariffs faced into RSA by
India
Brazil
100.0
5.8
2.0
2.0
Other grains
70.0
5.4
9.1
30.1
Vegetables/fruit
45.4
11.5
4.3
0.5
Oil seeds
31.3
5.2
8.6
0.0
Plant fibres
5.0
8.1
5.6
14.5
Other crops
32.3
9.1
6.2
14.0
Cattle
31.0
0.0
0.0
0.0
0.0
4.5
0.0
0.0
15.0
9.5
0.0
0.0
Beef/sheep meat
35.0
11.2
20.0
17.4
Other meat
30.0
7.5
23.5
8.8
Vegetable oils
51.6
11.0
3.0
9.8
Dairy products
30.0
15.5
53.3
0.0
Processed rice
75.0
14.3
0.0
0.0
Sugar
60.0
17.5
0.0
0.0
Other food
39.6
13.4
12.6
17.0
Beverage, tobacco
41.0
21.7
31.4
145.8
Fish
30.0
10.3
5.9
0.0
5.7
8.2
0.3
0.0
15.0
0.0
0.0
0.0
8.6
4.5
0.1
0.0
Other agricultural products
Wool
Secondary and natural resources
Forestry
Coal/oil/gas
Other minerals
29
Tariffs faced, South Africa into
Primary
India
Brazil
Tariffs faced into RSA by
India
Brazil
Manufacturing
Textiles
16.1
16.7
21.5
15.1
Wearing apparel
15.0
17.4
39.1
11.1
Leather goods
23.8
16.3
12.4
13.2
Wood products
13.0
11.3
16.8
8.4
Paper products
10.5
13.6
4.8
5.4
Petroleum, coal/oil/gas
products
Chem rubber plastic
12.9
2.5
4.2
2.0
14.9
9.6
3.6
3.0
Mineral products (oth)
15.0
11.6
7.8
13.7
Iron steel
19.9
11.1
3.9
1.2
Other metals
15.0
6.6
4.0
0.7
Other metal products
15.1
17.3
9.9
8.1
Motor vehicles, etc.
16.3
17.5
15.9
23.6
Other transport equipment
6.8
1.2
0.1
0.0
Electrical goods
2.1
10.0
8.6
18.9
Other machinery
12.7
14.8
4.2
4.6
Other manufactures
15.0
16.3
5.5
13.3
Results: the implications of the IBSA FTA
The big picture
Table 13 shows the changes in welfare from the FTA assuming a complete 100 percent
reduction in merchandise tariffs, with the data expressed in US$ million as one-off increases
in annual welfare at the assessed end point of 2015. South Africa’s gains are $1.448 million,
a figure very close to Brazil’s $1,514 million but nearly 50 percent above India’s
$1,022 million. Note that both Botswana and Rest of SACU actually lose in welfare terms,
and this results mainly (a) because there is little reported trade between these SACU
partners12 and either India or Brazil in order to generate increases (with this reflected in the
terms of trade losses) and (b) there is a considerable downside for these countries in that
tariff revenue is being lost for the SACU pool13. The biggest loser in dollar terms is the EU,
12
A well-known weakness of a trade model is that it is not able to provide any insights into potential
new areas of trade (or changes in trade from a very low base) that may develop from an FTA, for
example, and this development of new trade (or increased trade from that minimal base) has been
shown to be an important part of the potential gains from an FTA.
13
We note that this revenue loss is incompletely covered in our version of the model and therefore the
welfare losses may represent an underestimate of the potential losses to the BLNS given their reliance
30
with all other countries/regions except Nigeria (NGA) losing. The gains to South Africa are
spread across the contributing factors of increased allocative efficiency ($338 million),
employment ($173 million), capital stock ($467 million) and the terms of trade gains from
better relative prices between exports and imports. Note in particular that, for the labour
markets, South Africa has the highest total contribution for any of the countries shown – this
is because of the labour market closure which is a function of the available unskilled labour
proxied by the unemployment rates. This aspect of the results will be discussed in more
detail later in the chapter.
Table 13 Change in welfare (EV of income) due to IBSA, $ million at 2015
Allocative
Total
Term of
Efficiency
Labour
Capital
Trade
1,448
338
173
467
470
Botswana
-18
1
-1
-5
-14
Rest SACU
-19
0
-2
-5
-12
40
3
1
6
30
-25
-24
1
-26
24
EU
-1,295
-412
-40
-435
-407
US
-342
-111
-33
-51
-147
India
1,022
-274
37
1,495
-236
China
-262
-31
-11
-88
-132
Brazil
1,514
411
72
574
457
Japan
-287
-68
-14
-57
-147
Rest of World
-925
-356
-42
-638
111
850
-524
139
1,237
-3
South Africa
Nigeria
Rest Africa
Total
Source: GTAP results
In further examining the GTAP results we are able to decompose the results to find that:
•
South Africa’s gains of $1,448 million overwhelmingly come from gains through better
access into India of $1,306 million, followed a long way back by better access into
Brazil of $110 million, while Brazil’s access into South Africa results in gains to
South Africa of $51 million. There are losses to South Africa from India’s enhanced
access into both Botswana and Rest of SACU of $8 million and $76 million
respectively.
on the SACU tariff pool for total government revenues. For an in-depth discussion of this aspect of
South African/SACU trade policy, see Sandrey et al. (2006) and Sandrey (2006b).
31
•
India’s gains of $1,022 million are dominated by gains of $714 million from increased
imports from South Africa into India. This story is ‘pure gold’ and will be discussed
later. India also gains $177 million from better market access into South Africa and
$512 million from better access into Brazil (and $18 million from better access into
Botswana), but loses some $400 million from better Brazilian access into India (in
direct contrast from gaining from South African penetration of its market).
•
The story for Brazil is different in that most ($1,225 million) of its gains of $1,514 million
result from better market access into India. South Africa is less important as far as
Brazil is concerned, although it does gain $220 million from better access into
South Africa. Other than gains of $63 million from India gaining access to Brazil and a
lesser $8 million from South Africa gaining better access to India, the other values
(including the impact of South African access to Brazil) are close to zero.
•
The EU loses some $1,295 million, with $4116 million of this as a result of South Africa
displacing the EU in India (in the sectors where South Africa makes gains into India)
and another $640 million from Brazil displacing it in the same Indian marketplace. It
also loses another $63 million and $87 million as firstly India and then Brazil displace it
(the EU) in the South African market (recall that this is post-TDCA).
•
Most ($224 million) of China’s $262 million loss results from Brazil displacing it into
India, although it also loses $18 million and a lesser $10 million as Brazil and India
respectively gain better access to South Africa (note that the latter figure is small
despite India making a large increase in clothing trade into South Africa).
•
The Rest of the World (RoW) loses mainly because of Brazil displacing it in the Indian
market.
•
Botswana’s losses are mostly from a $35 million loss from South Africa’s gaining better
access into India, increasing South Africa’s export prices, which result in a term of
trade loss for Botswana, thus changing the relative position for Botswana – although it
gains some $154 million in compensation as India obtains better market access into
Botswana.
•
For XSC (Rest of SACU, or Namibia, Lesotho and Swaziland combined) the
$19 million loss is from a loss of $21 million as South Africa gains increased market
access into India increasing its export prices resulting in a terms of trade loss also for
the Rest of SACU.
•
For the total, GTAP is showing that IBSA is welfare-enhancing for the world, as world
welfare increases by $850 million (and, as shown in Table 13, this is mostly from
32
increased capital stock in that the IBSA FTA increases regional income, which
increases savings/investments, resulting in increased global capital stocks).14
Before moving on it behoves us to say more about the ‘golden story’. As mentioned above,
both South African and Indian gains are dominated by gains from allowing better access from
South Africa into India. Examining both the GTAP output and bilateral trade flows in detail
allows us to trace this through this effect to one of allowing South African gold to enter India
duty-free as distinct from the 15 percent duty that it would normally carry. Looking at Indian
import data for 2005, we find that gold is the second main import behind fuel, with an import
value of $11.74 billion. Of this, some $5.97 billion is from Switzerland 15 , followed by
$2.13 billion from South Africa and $1.84 billion from Australia. Thus, given the huge
15 percent advantage to South Africa, there is a doubling in trade here, and, as the Indian
Input-Output (I-O) table that drives GTAP shows, this import has a large flow-on effect
through Indian production and, consequently, jewellery exports. As jewellery (HS 7113) is the
fourth main export globally from India with a 3.19 percent share (diamonds are the main
export with a 11.12 percent share) this gives a massive boost to Indian trade in jewellery.
Overall, the gains for India of $1,022 million are dominated by gains of $660 million from the
metals not elsewhere classified (n.e.c.) sector, and examining the Indian capital flows in
detail confirms this as the main point of interest in the IBSA.
From a policy perspective, this is a win-win situation for both South Africa and India,
and if trade negotiators have to focus on one key point, lowering the tariffs for South
African gold into India is that one point. Here is a meeting of one of the world’s main gold
producers with the major jewellery producing and exporting nation. In the GTAP database
and model, South Africa is exporting the aggregated GTAP commodity non-ferrous metals
(NFM, metal n.e.c.) to India which is mainly gold metal in some form. The South African NFM
industry uses minerals mined in South Africa as an input into its production of metals which
come from the GTAP mining sector OME (minerals n.e.c.). In the analysis undertaken in this
paper we have not extended the GTAP analysis to review exactly what natural resources
14
With the creation of the IBSA FTA regional income will initially increase through possible allocative
efficiency, terms of trade and increased employment gains in the economy. This increased income
causes savings and investments to increase in the economy, which result in an expansion of the
capital stock. This increase in capital stock causes regional income to increase even further. The
model finds a new equilibrium capital stock where the marginal return to capital is equal to the
depreciation rate of capital. At this point, regional income would fall if the economy expands its capital
stock further.
15
We would note in passing that there are few people actually digging gold out of the ground in
Switzerland, and as a leading global gold producer the effective transfer of gold from South Africa to
India is likely to be higher than Indian imports show. During 2005 South Africa was the leading
individual country in gold production with an 11.8 percent share.
See: http://www.goldsheetlinks.com/production.htm.
33
(minerals) other than gold are being mined in South Africa. Gold certainly dominates,
although other minerals such as platinum are included which do not feature in either South
African exports to India or Indian imports from South Africa. In the model, the amount of
natural resources being mined is fixed exogenously, so there can be no change in the
quantity, and this model closure does not make any distinction as to what type of mineral is
being mined - gold or iron ore in the OME GTAP sector. Therefore the expanding NFM
(metals n.e.c.) sector in South Africa is pushing up the price of natural resources (minerals),
and increasing the cost of production in other South African sectors/commodities like the
motor vehicle sector. If South Africa chose to mine a larger quantity it would lower the price
of natural resources (minerals) in South Africa, lowering the production cost in both the NFM
sector but also other sectors which use OME (machinery and equipment n.e.c.) as inputs into
their production (e.g. motor vehicles). This would of course increase the production of NFM
and exports to India even further. This necessitates changing the structures of the GTAP
model that we are using, and, even then, changing the amount of mining done opens up
further questions about the maximisation of resources in the ground through time: more
minerals from the mine today or wait until some time in the future? We consider that this
wider analysis is beyond the scope of the present study.
Table 14 expands upon the welfare gains to show on the left-hand side what the actual
changes are, while the right-hand side provides some insights into where the contributions to
these welfare changes are coming from. In column 3 (real GDP) South Africa gains by some
0.90 percent, while both Botswana and the Rest of SACU lose by around 0.10 percent (both
India and Brazil gain by around 0.2%). These results flow in large part from Table 13 above,
as the allocative efficiency, labour and capital contributions are essential components of the
real GDP changes with resources being better used within the economy. South Africa gains
more in percentage terms as, although the welfare gains are similar in dollar values, the
relative sizes of the three main economies are not. The same applies to the terms of trade
(TOT), expressed as dollar millions above in Table 13 but in percentage terms in Table 14
below.
34
Table 14: Percentage change in terms of trade, real GDP and factor income, 2015
TOT
RSA
Total
Of which contributions come from
Real
factor
Unskilled
Skilled
GDP
income
labour
labour
Land
Natural
Capital
resources
1.04
0.90
2.05
0.03
0.72
0.34
0.65
0.31
Botswana
-0.49
-0.09
-0.22
0.01
-0.09
-0.05
-0.11
0.01
RSACU
-0.28
-0.14
-0.12
0.07
-0.10
-0.05
-0.06
0.02
India
-0.20
0.17
0.33
-0.01
0.11
0.03
0.17
0.03
Brazil
0.53
0.21
0.71
0.13
0.20
0.13
0.26
0.00
Source: GTAP output
On the right-hand side of Table 14 the relative contributions to total factor income are shown,
and indeed these must add to equate with the total factor income percentages shown. Thus,
for South Africa’s 2.05 percentage increases in total factor income, 0.03 percent is derived
from better land use, 0.72 percent from enhanced unskilled labour, 0.34 percent from skilled
labour, 0.65 percent from capital, and the final 0.31 percent from the natural resource base.
Note that for land, skilled labour and natural resources, the quantities are fixed in the GTAP
model, so the increases result from price increases as their values are bid up, while for both
unskilled labour and capital where the quantities are not fixed, there is both a price and a
quantity effect. The unskilled labour issue will be examined in more detail later under
alternative model closures, as these policy options for (a) holding unskilled labour fixed and
raising the wage rate, (b) holding the wage rate fixed and increasing the people in
employment and (c) some intermediate position, are crucial policy questions for South Africa.
Currently we are using a closure of (c) as discussed earlier.
Changes in trade flows
The details of changes to South Africa’s trading patterns are shown in Annex A, with exports
in Annex Table A1 and imports in Annex Table A2 respectively. These tables split the GTAP
sectors into primary and secondary agriculture, natural resource and manufacturing before
displaying (a) the AVE or the initial pre-FTA average ad valorem tariff facing either South
African exports in the partner market for exports or partner tariffs into South Africa for
imports, and (b) the change in either South African exports or imports into or from the
respective country/regions in response to reducing these border tariffs to zero. These trade
flow changes are given in both US dollar values and percentage changes from the base to
put them in perspective. The markets for exports from and sources of imports into South
Africa are provided as India, Brazil, the Rest of SACU (with Botswana and XSA combined)
35
and the Rest of the World with all others combined rather than the RoW used in the model.
There are also two columns on the right-hand side showing the total overall changes by
value and percentage change.
Table 15 shows the aggregate overall changes to trade flows for the partner countries in
2015, expressed as percentage changes for both exports and imports and then in US$
million for the trade balance. The clear gainer is South Africa, with increased exports
globally, once all markets are accounted for, of $253 million, while India and Brazil face
losses of some $389 million and $59 million respectively as increased exports are less than
the import changes. Botswana has a deteriorating trade balance of $4 million16, while the
change for the Rest of SACU, while negative, registers on the table in terms of a trade
balance of $1 million.
Table 15: Percentage change in the quantity of total imp\exp and trade balance, 2015
South Africa
Botswana
RSACU
Exports
2.1
-0.2
-0.2
2.3
1.0
Imports
3.7
-0.7
-0.5
2.9
1.6
253
-4
1
-389
-59
Trade balance US$ million
India
Brazil
Source: GTAP results
Table 16 extends this analysis to look at the agricultural sector changes. Recall that the
factors of production of land, skilled labour and natural resources are fixed, while unskilled
labour and capital can increase in both price and quantity. We can see that for all SACU
partners the impacts upon agricultural factor income are positive, and in particular, for both
South Africa and RSACU. Land prices are increasing here in all SACU, and unskilled labour
prices are increasing in South Africa and RSACU. Thus, IBSA benefits the SACU agricultural
sector. Brazilian land prices are increasing by nearly two percent (as a result of better
access into India rather than South Africa), while India’s marginally declines.
16
Note that Table 13 refers to a deteriorating terms of trade (ToT) for Botswana, while Table 15 shows
the actual trade flows. They are different concepts, as ToT is the relative indicator price of exports and
imports.
36
Table 16 Percentage change in primary agricultural factor income, 2015
Of which contributions come from
Agricultural
factor
income
1.16
Land
0.57
Unskilled
labour
0.45
Skilled
labour
0.01
Capital
0.14
Botswana
0.24
0.22
0.00
0.00
0.02
RSACU
0.68
0.56
0.05
0.00
0.07
India
0.02
-0.03
0.05
0.00
0.00
Brazil
2.48
1.87
0.24
0.01
0.36
South Africa
The specific sector results
This section will discuss the production, trade and relative prices changes in the GTAP
sectors. The following series of tables examining the changes to South African trade
flows is largely drawn from the Annex tables.
We will start with the agricultural and natural resource sectors to keep the analysis
manageable, and these are shown in Table 17. Keep in mind that these are percentage
changes, and some of these may be off low bases. Further analysis of the results (see
Annex Table A1) shows that in the agricultural sector, South Africa increases exports to
India by $182 million and Brazil by a much lesser $7 million, but overall some $140 million of
this is at the expense of exports to other SACU ($2 million less) and the Rest of the World.
Thus, the total increase is only $44 million. Of the total increase, most is in vegetable oils and
fats ($68 million) and wool ($30 million) where big increases to India take place, while there
are global reductions of $16 million each in exports of (a) vegetables, fruit and nuts and (b)
other food products. For imports, there is a similar but slightly larger overall increase of
$92 million, driven mostly by increased imports from Brazil of $76 million (other crops, other
meats and vegetable oils and fats). In the natural resources sector, there is an increase of
$189 million globally, dominated by coal, oil, gas products of $178 million as the percentage
changes shown in coal oil gas are off large bases. Exports of coal/oil/gas to India increase
by some $824 million (over 500%) as the duty rates of 15 percent are slashed, and this
necessitates imports into South Africa from the Rest of the World as South Africa’s relative
price increases domestically 17 . In summary, there is little action in either exports or
imports in the agricultural sectors, and therefore limited production changes, while
only coal/oil/gas products show increased trade in the natural resources sectors.
17
Once again the quantity of coal/oil/gas being extracted from the mines in South Africa is fixed
exogenously in the model, therefore the increased demand of coal/oil/gas by India increases the
South African coal/oil/gas price but with no change in the quantity mined.
37
The interpretation of Table 17 follows, using wheat as an example. The table shows that real
wheat prices actually increase by 0.7 percent (right-hand column). This, however, does not
seem to be enough to attract resources into wheat production as other sectors (in both
agriculture and non-agricultural market access) become relatively more profitable.
Consequently, imports of wheat rise by 1.9 percent and exports decline by 3.0 percent. This
leads to a final decline of 1.4 percent in South African wheat production as a result of IBSA.
Table 17: Relative changes in the South African agricultural and natural resource
sectors, %
Paddy rice
Change
Change
Change
Change in
in quantity
quantity of
quantity of
Real price
output
exports
imports
of output
0.3
2.3
2.4
1.1
Wheat
-1.4
-3.0
1.9
0.7
Cereals grains
-0.2
-1.5
6.2
0.9
Vegetables, fruit, nuts
-0.9
-1.7
1.2
0.8
0.9
-2.4
5.5
1.1
Sugar cane & beet
-0.9
-3.3
0.2
0.8
Plant based fibres
-1.7
-1.9
1.1
0.5
Crops nec
0.4
2.9
4.5
1.0
Bovine cattle
0.6
-2.5
1.9
1.0
Animal products nec
0.1
-1.9
0.6
0.9
Raw milk
0.2
-6.2
3.4
0.9
14.1
18.8
10.6
3.4
Fishery
0.0
-2.8
1.8
1.3
Forestry
0.3
2.3
1.9
1.2
Coal/oil/gas products
0.0
1.5
2.9
1.5
Minerals nec
1.0
-4.2
6.5
3.1
Beef
0.5
-2.8
4.2
0.9
-1.1
-3.8
11.3
0.8
7.8
67.0
7.4
0.7
Milk
-0.2
-4.0
3.7
0.8
Processed rice
-0.3
-1.5
1.8
0.8
Sugar products
-1.3
-2.8
2.0
0.8
Other foods
0.1
-2.1
3.3
0.8
Beverages, tobacco
0.5
-1.1
1.8
0.7
Oil seed
Wool, silk cocoons
Other meat
Vegetable oils
Source: GTAP results
38
Of particular interest from Table 17 is the sugar sector, as Brazil and India, and to a lesser
extent South Africa, are major global players. South African production declines by
1.3 percent as resources are displaced to slightly more profitable sectors (in both agricultural
but more particularly non-agricultural sectors as we shall see next). Analysis of the detailed
GTAP trade data results shows imperceptible changes to the trade flows here despite the
high sugar tariffs into India. However, South Africa has no initial trade with India in the GTAP
database, so the removal of the 60 percent tariff may have a large percentage increase in
the quantity of exports to India, but, as the base is nearly zero, the change in value of exports
is minimal. Our earlier trade chilling analysis suggested that this sector held potential for
South African exports to India, and this highlights the need to undertake trade analysis
‘outside of the model’.
We now turn our attention to the manufacturing sectors, where the account is completely
different and there is a large response for South Africa. This is shown in Table 18, where the
‘golden story’ of the large response in other metals dominates. Note that Table 18 is best
read in conjunction with Table 19, which shows the bilateral changes to South African
manufacturing exports as a result of IBSA. While the same data is available for South
African imports, it is not shown but rather discussed for the main sectors impacted by
imports.
Table 18: Relative changes in the South African manufacturing sectors, %
Textiles
Wearing apparel
Leather
Wood products
Paper products
Petrol, coal, oil, gas products
Chemicals, rubber, plastic
Other mineral products
Iron/steel
Other metals (gold)
Metal products
Motor vehicles
Other transport
Electronics
Machinery equipment
Other manufactures
Services
Source: GTAP results
Change
in quantity
output
-1.1
-11.0
-3.0
-1.6
-0.1
0.3
1.6
-1.2
-1.4
15.7
-1.8
-1.6
-2.4
-5.3
-2.7
-6.7
1.1
Change
quantity of
exports
2.5
-1.5
-2.5
-4.4
-1.4
-1.1
6.2
-2.5
-0.4
17.2
-4.1
-1.6
-4.1
-7.5
-4.6
-7.7
-3.5
Change
quantity of
imports
6.0
16.0
6.3
3.5
2.9
2.9
3.5
4.0
2.6
14.2
7.3
4.3
2.1
1.5
3.0
3.4
3.1
Change in
real price
of output
0.6
0.0
0.6
0.9
0.8
0.7
0.9
1.0
1.1
1.1
0.8
0.4
0.8
1.0
0.9
1.1
1.0
39
Table 19 shows that manufacturing exports to India increase by a huge $2,752 million, with
$2,210 million of this in the metals n.e.c. (mostly gold) sector as exports to India double.
Other big gainers are the chemical, plastics and rubber sector where exports to India
increase by $312 million. The lesser gains of $238 million to Brazil are more dispersed, but
concentrated in the metals n.e.c. sectors again: chemicals, rubber and plastics; machinery
and equipment; and textiles (possibly wool?). The big increases in exports to India are
responsible for South Africa’s increasing its global exports of manufacturing goods by
$1,617 million despite declines to other destinations. Note on the right-hand column of Table
18 (and also Table 17) that the resource price in all sectors is increasing by at least a small
positive amount and most around one percent.
Table 19: Changes in South African bilateral exports for manufacturing products,
$ million and %
To India
AVE
To Brazil
change
% change
value
quantity
AVE
change
% change
value
quantity
tariff US$ million of exports tariff US$ million of exports
Manufacturing
Textiles
16.1
13.6
194
16.7
22
198
Wearing apparel
15.0
0.3
183
17.4
0
209
Leather products
23.8
4.8
418
16.3
1
219
Wood products
13.0
0.3
120
11.3
2
99
Paper products, publishing
10.5
20.5
71
13.6
9
104
Petroleum, coal oil gas products
12.9
19.6
62
2.5
0
8
Chemical, rubber, plastic
14.9
311.6
133
9.6
63
70
Mineral products n.e.c.
15.0
5.9
111
11.6
3
77
Ferrous metal
19.9
91.7
170
11.1
18
71
Metals n.e.c.
15.0
2,209.8
99
6.6
70
50
Metal products
15.1
8.4
169
17.3
13
194
Motor vehicles and parts
16.3
10.6
104
17.5
15
142
Transport equipment n.e.c.
6.8
0.2
66
1.2
4
3
Electronic equipment
2.1
0.2
11
10.0
0
112
Machinery and equipment
12.7
48.3
144
14.8
17
181
Manufactures n.e.c.
15.0
6.0
162
16.3
1
185
Total Manufacturing
Source: GTAP results
2,751.7
238
40
Not shown is the companion table to Table 19 for imports. Here the two big changes are in
(a) clothing imports of an increased $257 million from India and (b) increased imports of
$240 million in motor vehicle imports from Brazil. If we look at Table 18 we see that clothing
production in South Africa declines by a large 11.0 percent from IBSA, and it is Indian
imports that are driving this reduction. Not shown is that imports into South Africa from SACU
are also displaced by India, (and India is making inroads into SACU) and this is the largest
overall change in that trade profile. These changes in the clothing sector resulting in an
overall decline for South Africa are a lesser replication of the results from a China FTA
as discussed in Jensen and Sandrey (2006), and confirm the possible expansion of
Indian imports into South Africa as discussed in Sandrey (2006c) in response to the
restrictions recently placed upon Chinese imports into South Africa.
There are several factors at play with respect to the South African motor vehicle (MHV)
sector. Global imports increase by $151 million, mostly through increases from Brazil
($240 million) and India ($46 million); this increase is in response to South African tariffs on
current trade of 23.6 and 15.9 percent for Brazil and India respectively falling to zero. This
new IBSA trade displaces some $131 million in imports from the Rest of the World.
Conversely, since the initial export values to India and Brazil were low, there are limited
compensating exports to India of $11 million and to Brazil of $15 million that do not match the
declines of $38 million to the Rest of the World. Overall production in the motor vehicle
sector declines by 1.6 percent, as shown in Table 19. Importantly, the main cause of this 1.6
percent decline (and of course declines in the other manufacturing sectors as well) is the
increased export (production) of NFM to India pushing up the price of the intermediate inputs
(especially OME) and endowments (labour capital) into the MHV sector which then reduces
MVH production by 1.3 percent. This leaves a reduction of a lesser 0.3 percent by other
effects. This increased price of the intermediate inputs and endowments in the MVH sector
increasing production prices in South Africa also explains the reduction in exports to the Rest
of the World due to increased production costs. This does not necessarily mean that the
South African MVH industry is less efficient than India’s or Brazil’s, but rather that the
industry now has to pay a higher price domestically for inputs.
The labour market closure and alternative scenarios
We believe that this particular section of the GTAP analysis of IBSA potentially makes an
important contribution to the nexus between trade policy and welfare redistribution in
South Africa. The results in this paper are driven from the labour market assumption as
displayed in rows (B) in Table 20 below, with the mathematical derivation of the equation
41
shown in an annex to Chapter 4 to ensure readers that it is both mathematically correct and
economically sensible.
The simulation results as presented above are those given in section (B) of Table 20, where
the welfare gains for South Africa are some $1,448 million and the increased real GDP is
0.90 percent. Not shown earlier is that, all other things held constant, IBSA will contribute a
further 0.65 percent to South Africa’s CPI. This, of course, has an impact upon the whole
economy and results in South Africa losing some competitive advantage globally. The key
employment issue is that shown in the centre of the table under ZAF, where the South
African story on unskilled labour is told (recall that the supply of skilled labour is held fixed).
Under this closure whereby the unskilled labour supply is a function of the unemployment
rate, the employment of unskilled labour increases by 0.46 percent and the real wage rate by
1.27 percent. To the right of this we see that both employment and the wage rate decline in
both Botswana and Rest of SACU, while employment increases only marginally in India and
Brazil – but the wage rate increases by 0.38 percent in India and 0.52 percent in Brazil as
there is much less reported unemployment in these two countries.
Table 20: Unskilled labour market closure, percentage change employment/real wage
South Africa
EV
QGDP
CPI
US$ million
%
%
1,145
0.66 0.693 Fixed
0.90
(B)
2,360
(D)
0.65 U
(1-U)
1.62
0.52 R. Wage
CPI
(C)
3,006
ZAF
employm
(A)
1,448
Rest of
2.13
0.43 Fixed
R. Wage
Botswana SACU
Employment
0.00
0.00
Real wage
1.52
-0.21
Employment
0.46
-0.052
Real wage
1.27
-0.17
-0.15 0.377 0.516
1.847
-0.467
-0.51 0.336 0.402
Real wage
0.52
0.07
0.17 0.102 0.339
Employment
2.83
-0.24
Real wage
0.00
0.00
Employment
0.00
India Brazil
0.00
0.00
-0.21 0.393 0.541
-0.08 0.017
-0.12 0.452
0.00
0.00
0.05
1.08
0.00
Source: GTAP results based upon alternative closure assumptions
Section (A) in Table 20 shows what may be thought of as a general COSATU position
seeking to protect those already in employment. In this extreme position the level of
employment is fixed and all adjustments must take place within the wage rate. This is, as
42
expected, good for those who are employed, as their wage rate increases by 1.52 percent. It
is not so good for either those not employed or the economy, as the welfare gains reduce to
$1,145 million and GDP is a lesser 0.666 percent higher while the CPI increases to
0.69 percent.
At the other extreme we have section (D), where the real wage is fixed and all adjustments
must come through the number of unskilled persons employed. Here the results are
striking: employment is up by 2.83 percent, welfare more than doubles to $3,003 million or
2.13 percent of real GDP, and inflation is a significantly lower 0.43 percent. This is a dramatic
result which clearly highlights that if South Africa is serious about increasing both welfare and
employment in the economy, then the more policies move towards creating jobs rather than
rewarding those actually in employment is a superior option for policy makers.
This is confirmed by alternative (C), where the real wage rate is pegged to the CPI
(recognising of course that the CPI is itself a function of the labour market closure). Here the
welfare gains are some $2,360 million or 1.62 percent of real GDP with the real wage
increase set at the inflation rate of 0.52 percent. The employment increase of 1.85 percent is
similarly around a half-way position between the closure that we are using and the extreme
position of fixing the real wage.
Other sensitivity scenarios
The 50 percent reduction
Another scenario was undertaken to simulate an across-the-board 50 percent reduction in all
IBSA tariffs to assess one variant of a less than comprehensive FTA.
The welfare results for South Africa are now $625 million (43% of the original), while those
for India are also $625 million (a greater 62% of the original), but Brazil’s gains shrink to
$444 million or just 30 percent of the original. For South Africa, all the gains/losses are of a
similar magnitude to the 43 percent overall figure (including the main gain from better access
into India) except for the $3 million loss from better Indian access into South Africa which
now turns into a modest $19 million gain for South Africa (although the Indian gains from
better access into South Africa remain at 52 percent of the original). At the sector level the
reduction in South Africa’s clothing sector is now 2.95 percent, only 34 percent of the full
reduction of 8.7 percent, while the reduction in motor vehicle production is 0.66 percent or
43
41 percent of the original reduction. In the ‘golden sector’ South African production of other
metals is up by 6.80 percent or 44 percent of the increase from the full FTA.
For India, much of the relative gain over the full FTA is that Brazil now makes limited inroads
into India, and the large loss resulting from this is almost neutralised. And for the same
reason Brazil’s overall gains are reduced by more than a straight-line 50 percent as its gains
in the Indian market are sharply reduced from $1,216 million to $307 million with partial
access into India.
Post-Doha IBSA analysis
The general assumptions and results regarding a possible Doha Round outcome have been
outlined earlier in Chapters 5 and 6 for agriculture and non-agricultural market access
(NAMA) respectively. These results will not be duplicated here.
The general welfare changes to IBSA following a modest but successful Doha outcome are
shown below in Table 21. The welfare gains for South Africa are reduced by 11 percent from
$1,448 billion to $1,288 billion. A cursory analysis of the results shows that the ‘golden
sector’ output still increases, but by a lower 13.96 percent (15.5% before), but apparel
production declines by a lesser 6.8 percent (8.7% before), and vehicles are practically
unchanged with a decline of 1.53 percent (1.6% before). From the model we can see that in
agricultural products South Africa is deemed to be sheltering its sensitive/special products of
sugar and milk products in agriculture, and textiles/apparel and motor vehicles in the NAMA
sectors.
Table 21: IBSA, post- versus pre-Doha outcome EV in $ million and % difference
Country
South Africa
Initial IBSA
Post-Doha
Difference,
EV $ million
EV $ million
$ million
1,448
1,288
-$160m or 11%
Botswana
-18
-15
$3m or 19%
Rest of SACU
-19
-14
$5m or 26%
India
1,022
859
-$163m or 16%
Brazil
1,514
1,253
-$261m or 17%
Source: GTAP results
Thus, while all gains are lower, South Africa loses relatively less than India and Brazil, and
the other SACU partners are better off.
44
The IBSA Implications for the BLNS countries
Table 22 replicates the relevant parts of Table 13 with the results for the BLNS countries. As
discussed, Botswana is a country in its own right, but Lesotho, Namibia and Swaziland are
aggregated into one region (XSC) in the GTAP model. Thus, we are really unable to make
definitive statements about XSC as their economies are totally different. Nonetheless, we
can say that IBSA leads to welfare losses for both Botswana and XSC.
Table 22 BLNS Change in welfare (EV of income) due to IBSA, $ million at 2015
Allocative
Total
Term of
Efficiency
Labour
Capital
Trade
2 BWA
-18
1
-1
-5
-14
3 XSC
-19
0
-2
-5
-12
Source: GTAP results (Table 14)
In further examining the GTAP results we are able to decompose the results to find that:
•
Botswana’s losses are mostly from South Africa’s gaining better access into India,
increasing South Africa’s export prices, and resulting in a term of trade loss for
Botswana, thus changing the relative position for Botswana – although it gains some
compensation as India obtains better market access into Botswana.
•
For XSC, Rest of SACU or Namibia, Lesotho and Swaziland combined, the loss is
from South Africa’s gaining increased market access into India. This is increasing
South African export prices, resulting in terms of trade loss also for the Rest of SACU.
As for Brazil, there is a limited effect from having Brazil in the FTA.
Lesotho, Namibia and Swaziland (XSC)
For XSC, increased trade with India is the main issue, where exports increase by
$11.5 million in sugar products and $13.0 million overall. Elsewhere, exports increase to
Brazil by $1.6 million but decline to South Africa by $2.9 million and decrease by $3.0 million
to the Rest of the World. The main increases in exports to South Africa are in paper and
paper products (possibly wood pulp from Swaziland), NFM, and cattle, while the main
decreases in exports to South Africa are in textiles and wearing apparel. Overall, this
translates into increased exports of sugar (1.5%), NFM (3.8%) and wood pulp and paper
products (1.5%), with decreases of 3.4 percent in wearing apparel. Relating these and other
export changes back to dollar values, sugar increased by $7.5 million (mostly to India), pulp
45
and paper by $4.5 million (to South Africa), NFM by $3.2 million (to South Africa) and live
cattle18 by $1.3 million (to South Africa).
For imports, many sectors change marginally, with few sectors standing out.
Overall
however, imports increase by $3.8 million, with this hiding increases from India of
$37.8 million, Brazil of $4.1 million, but a large decline of $50.5 million from South Africa as
that source is displaced. The biggest change in value is in metal products where an increase
of $2.9 million takes place once increases to India of $17.2 million are set against declines to
South Africa of $11.8 million. Other increases are $1.1 million in petroleum and coal/oil/gas
products and $0.5 million in NFM (South Africa’s ‘golden’ sector), while losses take place in
reduced textile imports of $1.8 million as India displaces some of the domestic apparel
production. Slight increases in agricultural prices result in increases of $0.2 million in both
wheat (from the US) and fruit and vegetables (from South Africa).
The end result of the trade changes is that production of wearing apparel declines by
3.4 percent in the face of increased Indian competition while sugar production increases by
1.4 percent and non-ferrous metals by 2.5 percent.
Botswana
Here the account is one of less direct change in exports, with an overall loss of some
$9.2 million. Overall, this derives from three sectors: clothing ($4.9 million), vehicles and their
parts ($3.3 million), and, surprisingly, as resources move marginally into other sectors and
Botswana’s costs increase very marginally – the manufactures n.e.c. (down $5.3 million), a
sector that contains diamonds and makes up nearly three-quarters of Botswana’s global
exports. For clothing and vehicles and parts, the exports to South Africa reduce, while for
diamonds the reduction is to the almost exclusive EU market. Changes in the agricultural
sector are minor, but include an increase of $0.65 million in milk exports to India and around
$0.3 million in beef and other meats to South Africa and India respectively.
For imports, the picture is that overall increases of $44.3 million from India and $1.6 million
from Brazil largely displace South African imports (down by $49.8 million) for a small overall
change of $5.1 million less than without IBSA. Much of this change takes place in the general
machinery and other equipment sector, where India and, to a lesser extent, Brazil displace
18
Cattle meat (beef) is a different sector from live cattle, and in CMT (beef) there is a reduction of
$0.66 million in imports overall as increased exports of $0.69 million to South Africa only partially
offset the reduction of $1.17 million in beef to the EU. This translates to a reduction of 0.7 percent in
production but an increase in price of 0.4 percent once all the relationships are cleared.
46
South African imports. India generally makes some inroads into most of the manufacturing
sectors, with this at the expense of not only South Africa but also the EU, China and the Rest
of Africa. There are no changes of interest to the agricultural imports other than a decline in
milk products from South Africa that match the increase to India (a change that results in no
extra production).
The production of both motor vehicles and their parts and clothing is down (by 4.3% and
14.6% respectively). Very modest increases are seen in most other sectors with the
exception of the crucial diamond sector (manufactures n.e.c. – although as mining production
is fixed in the model, this may be somewhat misleading in a similar manner that the gold
story in South Africa is complex, as discussed earlier).
References
Bouet, A., Decrex, Y., Fontagne, L., Jean, S. and Laborde, D. 2005. A consistent, advalorem equivalent measure of applied protection across the world: The MAcMap-HS6
database. CEPII No 2004 – 22 December (updated September 2005). [Online]. Available:
http://www.cepii.fr/anglaisgraph/workpap/pdf/2004/wp04-22.pdf.
Dimaranan, B.V. (ed.). 2005. Global trade, Assistance and production: the GTAP 6 Data
Base. Center for Global trade Analysis, Purdue University, West Lafayette.
Jensen, H.G. and Sandrey, R. 2006. A possible SACU/China Free Trade Agreement (FTA):
Implications for the South African manufacturing sector. tralac Working Paper No. 8.
Stellenbosch: US Printers.
Sandrey, R. 2006a. South African Merchandise Trade with India. tralac Working Paper No.
10. Stellenbosch: US Printers.
Sandrey, R. 2006b. Trade Creation and Trade Diversion Resulting from SACU trading
Agreements. tralac Working Paper No. 11. Stellenbosch: US Printers.
Sandrey, R. 2006c. The Trade and Economic Implications of the South African Restrictions
regime on imports of clothing from China. tralac Working Paper No. 16. Stellenbosch: US
Printers.
47
Sandrey, R., Matlanyane, A. and Maleleka, D. 2006. Trade Liberalisation: What exactly does
it mean for Lesotho? tralac Working Paper No. 7. Stellenbosch: US Printers.
Stern, M., and Flatters, F. 2005. Business Day. 7 November. [Online]. Available:
http://qed.econ.queensu.ca/faculty/flatters/writings/geekonomics_1.pdf.
Van Meijl, H., Van Rheenen, T., Tabeau, A. and Eickhout, B. 2006. The Impact of different
policy environments on agricultural land use in Europe. Agriculture Ecosystems &
Environment 114 (2006) 21 – 38.
Walmsley, T.L. 2006. A Baseline Scenario for Dynamic GTAP Model. Revised March 2006
for the GTAP 6 Database. Center for Global Trade Analysis, Purdue University, West
Lafayette.
48
Annex Table A1: South African Changes in Exports with IBSA FTA, Initial average tariff and changes in exports by $ million and percentage
AVE
tariff
Primary
1 pdr
2 wht
3 gro
4 v_f
5 osd
6 c_b
7 pfb
8 ocr
9 ctl
10 oap
11 rmk
12 wol
India
Change
in value
US$
million
% Change
quantity
of exports
AVE
tariff
Brazil
Change % Change
in value quantity
US$
of exports
million
Rest SACU (inc Botswana)
AVE
Change % Change
tariff
in
quantity
value of exports
US$
million
RoW
Change % Change
in value quantity
US$
of exports
$million
Total
Change % Change
in value quantity
US$
of exports
million
RoW
AVE
tariff
80.0
100.0
70.0
45.4
31.3
0.0
5.0
32.3
31.0
0.0
0.0
15.0
0.1
4.5
0.2
6.2
0.0
0.0
0.1
27.3
0.0
0.0
0.0
63.1
101.5
large
large
279
289
258
-4
25
460
183
-2
-6
294
11.1
5.8
5.4
11.5
5.2
9.5
8.1
9.1
0.0
4.5
0.0
9.5
0
0
0
0
0
0
0
4
0
0
0
0
5
178
57
14
48
58
61
48
70
-2
11
-2
142
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
0
0
0
0
0
0
0
0
0
0
0
-1
-1.9
-3.9
-0.6
-0.6
-0.9
-0.6
-1.7
-1.9
-1.5
-0.8
-5.6
-21.5
0
-10
-2
-23
-1
0
-1
-15
0
-2
0
-33
-88
-9.2
-4.8
-1.8
-2.2
-2.5
-3.8
-2.2
-5.2
-3.4
-2.1
-6.3
-32.9
0.1
-5.7
-1.8
-16.2
-1.0
0.0
-0.6
16.0
-0.4
-1.7
-0.4
29.6
17.9
2.3
-3.0
-1.5
-1.7
-2.4
-3.3
-1.9
2.9
-2.5
-1.9
-6.2
18.8
3.7
132.9
43.8
6.5
8.8
0.9
0.9
10.6
3.5
2.6
0.0
1.0
Secondary
17 cmt
18 omt
19 vol
20 mil
21 pcr
22 sgr
23 ofd
24 b_t
35.0
30.0
51.6
30.0
75.0
60.0
39.6
41.0
1.6
1.0
70.0
1.3
0.0
1.4
4.8
0.4
80.6
771
765
large
534
large
887
267
117
11.2
7.5
11.0
15.5
14.3
17.5
13.4
21.7
0
0
0
0
0
0
0
1
2
118
81
111
177
94
133
63
55
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
-1
0
-1
0
0
-2
0
-4
-3.0
-1.8
-1.1
-2.8
-0.6
-3.5
-1.5
-0.8
-3
-5
-2
-7
0
-12
-20
-4
-52
-5.6
-5.9
-3.2
-5.3
-3.4
-3.1
-2.8
-1.4
-1.6
-4.4
68.3
-6.7
-0.3
-10.3
-16.2
-2.9
26.0
-2.8
-3.8
67.0
-4.0
-1.5
-2.8
-2.1
-1.1
54.7
10.5
11.4
26.2
10.5
30.6
16.1
21.2
Natural resources
13 fish
14 for
15 coal oil gas
16 omn
30.0
5.7
15.0
8.6
0.0
9.7
823.2
9.9
842.9
88
26
549
10
10.3
8.2
0.0
4.5
0
0
-11
0
-11
24
42
-17
2
0.0
0.0
0.0
0.0
0
0
-1
0
0
-1.2
-1.6
-13.2
-1.8
-1
-4
-676
-41
-723
-2.9
-5.0
-17.3
-4.7
-1.5
5.6
135.0
-30.4
108.6
-2.8
2.3
1.5
-4.2
8.1
3.4
1.6
0.5
Manufacturing
25 tex
16.1
13.6
194
16.7
22
198
0.0
-7
-8.6
-14
-4.2
14.9
2.5
7.2
49
AVE
tariff
26 wap
27 lea
28 lum
29 ppp
30 p_c
31 crp
32 nmm
33 i_s
34 nfm
35 fmp
36 mvh
37 otn
38 ele
39 ome
40 omf
41 srv
Total
15.0
23.8
13.0
10.5
12.9
14.9
15.0
19.9
15.0
15.1
16.3
6.8
2.1
12.7
15.0
0.0
India
Change
in value
US$
million
0.3
4.8
0.3
20.5
19.6
311.6
5.9
91.7
2,209.8
8.4
10.6
0.2
0.2
48.3
6.0
2,751.7
-0.9
3,775.8
% Change
quantity
of exports
AVE
tariff
183
418
120
71
62
133
111
170
99
169
104
66
11
144
162
17.4
16.3
11.3
13.6
2.5
9.6
11.6
11.1
6.6
17.3
17.5
1.2
10.0
14.8
16.3
-3
0.0
Brazil
Change % Change
in value quantity
US$
of exports
million
0
1
2
9
0
63
3
18
70
13
15
4
0
17
1
238
-1
233
209
219
99
104
8
70
77
71
50
194
142
3
112
181
185
-3
Rest SACU (inc Botswana)
AVE
Change % Change
tariff
in
quantity
value of exports
US$
million
0.0
-5
-4.8
0.0
-2
-3.7
0.0
-2
-2.2
0.0
-2
-1.5
0.0
1
-0.4
0.0
-3
-1.4
0.0
-1
-1.9
0.0
-1
-2.1
0.0
-2
-11.9
0.0
-15
-7.9
0.0
-5
-1.7
0.0
-1
-4.9
0.0
-5
-6.4
0.0
-39
-6.1
0.0
-3
-5.8
-90
0.0
-2
-3.8
-97
RoW
Change % Change
in value quantity
US$
of exports
$million
0
-9
-33
-36
-35
-116
-15
-95
-521
-41
-58
-18
-17
-166
-111
-1283
-125
-2,271
0.3
-4.8
-5.0
-4.2
-2.1
-5.2
-5.2
-5.6
-8.2
-5.8
-2.4
-6.3
-8.3
-6.6
-8.2
-3.7
Total
Change % Change
in value quantity
US$
of exports
million
-3.9
-5.3
-33.2
-7.9
-14.2
255.3
-7.8
14.3
1,757.8
-34.2
-37.5
-15.2
-21.5
-138.5
-106.4
1,616.7
-128.6
1,640.6
RoW
AVE
tariff
-1.5
-2.5
-4.4
-1.4
-1.1
6.2
-2.5
-0.4
17.2
-4.1
-1.6
-4.1
-7.5
-4.6
-7.7
7.8
3.9
2.3
4.7
7.6
5.9
6.6
3.1
0.9
4.2
5.2
2.1
6.2
3.7
2.0
-3.5
2.1
0.0
50
Annex Table A2: South African Changes in Imports with IBSA FTA, Initial average tariff and changes in exports by $ million and percentage
AVE
tariff
Primary
1 pdr
2 wht
3 gro
4 v_f
5 osd
6 c_b
7 pfb
8 ocr
9 ctl
10 oap
11 rmk
12 wol
Secondary
17 cmt
18 omt
19 vol
20 mil
21 pcr
22 sgr
23 ofd
24 b_t
Natural resources
13 fish
14 for
15 coal oil gas
16 omn
India
Change
in value
US$
million
% Change
quantity
of imports
AVE
tariff
Brazil
Change % Change
in value quantity
US$
of imports
million
Rest SACU (inc Botswana)
Change % Change
in value quantity
AVE
US$
of imports
tariff
million
RoW
Change % Change
in
quantity
value of imports
US$
million
total
Change % Change
in value quantity
US$
of imports
million
RoW
AVE
tariff
0.0
2.0
9.1
4.3
8.6
0.0
5.6
6.2
0.0
0.0
0.0
0.0
0
0
0
0
0
0
0
0
0
0
0
0
1
2.3
21.1
19.6
18.4
60.8
1.4
26.3
43.5
3.2
0.5
2.8
11.8
0.0
2.0
30.1
0.5
0.0
0.0
14.5
14.0
0.0
0.0
0.0
0.0
0
0
4
0
0
0
3
22
0
0
0
0
30
-9.1
15.0
84.6
-0.3
-4.8
-3.6
77.0
112.3
-1.3
-1.8
-5.9
-45.3
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
0
0
0
0
0
0
0
2
0
0
0
2
0.5
0.9
-5.1
0.3
4.1
-2.1
-3.7
-5.2
1.9
0.2
0.8
7.5
0
3
-2
1
1
0
-2
-9
0
1
0
2
-6
2.7
1.9
-4.6
1.4
5.4
1.6
-3.3
-3.6
2.9
0.8
3.5
10.6
0
3
3
1
2
0
1
13
2
1
0
2
26
2.4
1.9
6.2
1.2
5.5
0.2
1.1
4.5
1.9
0.6
3.4
10.6
0.0
2.0
6.3
5.2
0.6
1.2
8.1
4.8
0.0
0.3
0.0
0.1
20.0
23.5
3.0
53.3
0.0
0.0
12.6
31.4
1
3
0
0
1
0
3
0
8
316.5
548.9
22.9
2216.2
1.5
2.5
61.8
89.4
17.4
8.8
9.8
0.0
0.0
0.0
17.0
145.8
1
15
16
0
0
0
12
1
46
236.4
86.9
69.8
-1.2
-0.9
-0.1
85.5
691.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
1
0
0
0
0
0
1
0
2
1.6
-6.9
-2.2
2.1
2.2
0.9
0.5
0.8
2
-4
-1
2
3
2
5
2
10
3.7
-4.6
-0.8
3.7
2.1
2.8
1.0
1.2
4
13
15
2
4
3
22
3
66
4.2
11.3
7.4
3.7
1.8
2.0
3.3
1.8
22.3
9.8
12.6
33.8
0.0
0.0
4.9
4.6
5.9
0.3
0.0
0.1
0
0
0
0
0
15.7
2.1
4.1
7.4
0.0
0.0
0.0
0.0
0
0
0
1
1
1.2
-1.0
2.4
6.3
0.0
0.0
0.0
0.0
0
0
1
1
2
1.6
1.1
1.3
5.8
0
0
177
8
185
2.0
2.5
2.9
6.6
1
0
178
11
189
1.8
1.9
2.9
6.5
5.0
51
AVE
tariff
Manufacturing
25 tex
26 wap
27 lea
28 lum
29 ppp
30 p_c
31 crp
32 nmm
33 i_s
34 nfm
35 fmp
36 mvh
37 otn
38 ele
39 ome
40 omf
41 srv
Total
India
Change
in value
US$
million
% Change
quantity
of imports
AVE
tariff
Brazil
Change % Change
in value quantity
of imports
US$
million
Rest SACU (inc Botswana)
Change % Change
in value quantity
AVE
of imports
US$
tariff
million
RoW
Change % Change
in
quantity
value of imports
US$
million
total
Change % Change
in value quantity
of imports
US$
million
RoW
AVE
tariff
21.5
39.1
12.4
16.8
4.8
4.2
3.6
7.8
3.9
4.0
9.9
15.9
0.1
8.6
4.2
5.5
96
257
30
4
2
1
36
10
15
3
72
46
0
12
87
25
697
301.8
445.8
139.8
187.8
35.3
21.3
30.1
55.6
24.7
68.2
99.0
119.8
3.2
110.7
44.6
50.2
15.1
11.1
13.2
8.4
5.4
2.0
3.0
13.7
1.2
0.7
8.1
23.6
0.0
18.9
4.6
13.3
6
0
15
5
2
0
11
22
1
1
2
240
0
9
27
3
344
163.6
1.5
147.2
71.8
36.9
9.9
22.7
108.8
4.5
18.9
68.6
210.7
-0.3
352.5
42.1
148.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
-3
-17
0
0
5
0
3
0
0
4
0
-4
0
0
0
-1
-14
-5.5
-51.6
-7.8
1.0
1.9
1.3
1.3
-0.2
-1.2
13.2
-4.6
-4.7
1.8
0.7
0.6
-1.3
-48
-172
-22
5
16
1
95
4
-4
116
-24
-131
22
22
55
-4
-69
-6.2
-52.1
-6.7
1.3
2.7
1.5
2.5
0.4
-1.1
13.9
-4.1
-4.1
2.3
0.8
1.0
-0.6
52
68
23
13
26
3
145
35
12
123
49
151
22
43
170
24
957
6.0
16.0
6.3
3.5
2.9
2.9
3.5
4.0
2.6
14.2
7.3
4.3
2.1
1.5
3.0
3.4
14.6
25.5
10.5
3.8
1.4
3.7
1.9
3.7
1.4
0.2
4.1
26.1
0.2
1.4
1.7
4.2
0.0
1
707
2.8
0.0
0
421
1.6
0.0
0
-8
3.1
148
267
3.1
150
1,388
3.1
3.7
0.0
Working Papers
2002
US safeguard measures on steel imports: specific implications
by Niel Joubert & Rian Geldenhuys.
WP 1/2002, April
A few reflections on Annex VI to the SADC Trade Protocol
by Jan Bohanes
WP 2/2002, August
Competition policy in a regional context: a SADC perspective on trade investment & competition issues
by Trudi Hartzenberg
WP 3/2002, November
Rules of Origin and Agriculture: some observations
by Hilton Zunckel
WP 4/2002, November
2003
A new anti-dumping regime for South Africa and SACU
by Stuart Clark & Gerhard Erasmus
WP 1/2003, May
Why build capacity in international trade law?
by Gerhard Erasmus
WP 2/2003, May
The regional integration facilitation forum: a simple answer to a complicated issue?
by Henry Mutai
WP 3/2003, July
The WTO GMO dispute
by Maxine Kennett
WP 4/2003, July
WTO accession
by Maxine Kennett
WP 5/2003, July
On the road to Cancun: a development perspective on EU trade policies
by Faizel Ismail
WP 6/2003, August
GATS: an update on the negotiations and developments of trade in services in SADC
by Adeline Tibakweitira
WP 7/2003, August
An evaluation of the capitals control debate: is there a case for controlling capital flows in the SACU-US free trade
agreement?
by Calvin Manduna
WP 8/2003, August
Non-smokers hooked on tobacco
by Calvin Manduna
WP 9/2003, August
Assessing the impact of trade liberalisation: the importance of policy complementarities and policy processes in a
SADC context
by Trudi Hartzenberg
WP 10/2003, October
An examination of regional trade agreements: a case study of the EC and the East African community
by Jeremy Everard John Streatfeild
WP 11/2003, October
Reforming the EU sugar regime: will Southern Africa still feature?
by Daniel Malzbender
WP 12/2003, October
2004
Complexities and inadequacies relating to certain provision of the General Agreement on Trade in Services
by Leon Steenkamp
WP 1/2004, March
Challenges posed by electronic commerce to the operation and implementation of the General Agreement on
Trade in Services
by Leon Steenkamp
WP 2/2004, March
Trade liberalisation and regional integration in SADC: policy synergies assessed in an industrial organisation
framework
by Martine Visser and Trudi Hartzenberg
WP 3/2004, March
Tanzania and AGOA: opportunities missed?
by Eckart Naumann and Linda Mtango
WP 4/2004, March
Rationale behind agricultural reform negotiations
by Hilton Zunkel
WP 5/2004, July
The impact of US-SACU FTA negotiations on Public Health in Southern Africa
by Tenu Avafia
WP 6/2004, November
Export Performance of the South African Automotive Industry
by Mareika Meyn
WP 7/2004 December
2005
Textiles and clothing: Reflections on the sector’s integration into the post-quota environment
by Eckart Naumann
WP 1/2005, March
Assessing the Causes of Sub-Saharan Africa's Declining Exports and Addressing Supply-Side Constraints
by Calvin Manduna
WP 2/2005, May
A Few Reflections on Annex VI to the SADC Trade Protocol
by Jan Bohanes
WP 3/2005, June
Tariff liberisation impacts of the EAC Customs Union in perspective
by Heinz - Michael Stahl
WP4/2005, August
Trade facilitation and the WTO: A critical analysis of proposals on trade facilitation and their implications for African
countries
by Gainmore Zanamwe
WP5/2005, September
An evaluation of the alternatives and possibilities for countries in sub-Saharan Africa to meet the sanitary
standards for entry into the international trade in animals and animal products
by Gideon K. Brückner
WP 6/2005, October
Dispute Settlement under COMESA
by Felix Maonera
WP7/2005, October
The Challenges Facing Least Developed Countries in the GATS Negotiations: A Case Study of Lesotho
by Calvin Manduna
WP8/2005. November
Rules of Origin under EPAs: Key Issues and New Directions
by Eckart Naumann
WP9/2005, December
Lesotho: Potential Export Diversification Study: July 2005
by Ron Sandrey, Adelaide Matlanyane, David Maleleka and Dirk Ernst van Seventer
WP10/2005, December
African Member States and the Negotiations on Dispute Settlement Reform in the World Trade Organization
by Clement Ng’ong’ola
WP11/2005, December
The ability of select sub-Saharan African countries to utilise TRIPs Flexibilities and Competition Law to ensure a
sustainable supply of essential medicines: A study of producing and importing countries
by Tenu Avafia, Jonathan Berger and Trudi Hartzenberg
WP12/2006, August
Intellectual Property, Education and Access to Knowledge in Southern Africa
by Andrew Rens, Achal Prabhala and Dick Kawooya
WP13/2006, August
The Genetic Use Restriction Technologies, Intellectual Property Rights and Sustainable Development in Eastern
and Southern Africa
by Patricia Kameri-Mbote and James Otieno-Odek
WP14/2006, August
2006
Agriculture and the World Trade Organization – 10 Years On
by Ron Sandrey
WP1/2006, January
Trade Liberalisation: What exactly does it mean for South Africa?
by Ron Sandrey
WP2/2006, March
South African merchandise trade with China
by Ron Sandrey
WP3/2006, March
The Multifibre Agreement – WTO Agreement on Textiles and Clothing
by Eckart Naumann
WP4/2006, April
The WTO – ten years on: trade and development
by Catherine Grant
WP5/2006, May
A review of the results of the 6th WTO Hong Kong Ministerial Conference – Considerations for African, Caribbean
and Pacific (ACP) Countries
by Calvin Manduna
WP6/2006, June
Trade Liberalisation: What exactly does it mean for Lesotho?
by Ron Sandrey , Adelaide Matlanyane and David Maleleka
WP7/2006, June
A possible SACU/China Free Trade Agreement (FTA): Implications for the South African manufacturing sector
by Hans Grinsted Jensen and Ron Sandrey
WP8/2006, July
Ecolabels and fish trade: Marine Stewardship Council certification and the South African hake industry
by Stefano Ponte
WP9/2006, August
South African Merchandise Trade with India
by Ron Sandrey
WP10/2006, August
Trade Creation and Trade Diversion Resulting from SACU trading Agreements
by Ron Sandrey
WP11/2006, August
The ability of select sub-Saharan African countries to utilise TRIPs Flexibilities and Competition Law to ensure a
sustainable supply of essential medicines: A study of producing and importing countries
byTenu Avafia, Jonathan Berger and Trudi Hartzenberg
WP12/2006, August
Intellectual Property, Education and Access to Knowledge in Southern Africa
by Andrew Rens, Achal Prabhala and Dick Kawooya
WP13/2006, August
The Genetic Use Restriction Technologies, Intellectual Property Rights and Sustainable Development in Eastern
and Southern Africa
by Patricia Kameri-Mbote and James Otieno-Odek
WP14/2006, August
Initiation of WTO Trade Disputes by the private sector – need for SADC/COMESA countries to develop national
mechanisms.
by Felix Maonera
WP15/2006, October
The Trade and Economic Implications of the South African Restrictions regime on imports of clothing from China
by Ron Sandrey
WP16/2006, October
The Memorandum of Understanding and quotas on textile and clothing imports from China: Who wins?
by Gustav Brink
WP17/2006, October
WTO and the Singapore Issues
by Ron Sandrey
WP 18/2006, November
How can South Africa exploit new opportunities in agricultural export markets? Lessons from the New Zealand
experience.
by Ron Sandrey & Nick Vink
WP 19/2006, November
Promoting agricultural trade and investment synergies between South Africa and other SADC Member countries.
by N. Vink, R. Sandrey, C.L. McCarthy & H.E. Zunckel
WP 20/2006, November
Proposed amendments to the anti-dumping regulations: are the amendments in order?
by Gustav Brink
WP 21/2006, November
2007
Examining the India, Brazil and South African (IBSA) Triangular Trading Relationship
Ron Sandrey and Hans Jensen
WP 1/2007, February
Government-Business Interface in dispute settlement: Lessons for SACU
Augustine Mandigora
WP2/2007, February
South Africa and Japan: towards a new trading relationship?
Ron Sandrey
WP3/2007, March
South African agriculture protection: how much policy space is there?
Ron Sandrey, Olubukola Oyewumi, Bonani Nyhodo and Nick Vink
WP4/2007, March
South African agriculture: a possible WTO outcome and FTA policy space - a modelling approach.
Ron Sandrey and Hans Jensen
WP5/2007, March
Revisiting the South African-China trading relationship
Ron Sandrey
WP6/2007, March
Safeguards in South Africa: What Lessons from the First Investigation?
Gustav Brink, G. 2007.
WP7/2007, May
South African Wine – An Industry in Ferment
Stefano Ponte and Joachim Ewert
WP8/2007, October
Governance in the Value Chain for South African Wine
Stefano Ponte
WP9/2007, October
Trade Briefs
2002
Cost sharing in international dispute settlement: some reflections in the context of SADC
by Jan Bohanes & Gerhard Erasmus.
TB 1/2002, July
Trade dispute between Zambia & Zimbabwe
by Tapiwa C. Gandidze.
TB 2/2002, August
2003
Non-tariff barriers: the reward of curtailed freedom
by Hilton Zunckel
TB 1/2003, February
The effects of globalization on negotiating tactics
by Gerhard Erasmus & Lee Padayachee
TB 2/2003, May
The US-SACU FTA : implications for wheat trade
by Hilton Zunckel
TB 3/2003, June
Memberships in multiple regional trading arrangements : legal implications for the conduct of trade negotiations
by Henry Mutai
TB 4/2003, August
2004
Apparel Trade and Quotas: Developments since AGOA’s inception and challenges ahead
by Eckart Naumann
TB 1/2004, March
Adequately boxing Africa in the debate on domestic support and export subsidies
by Hilton E Zunckel
TB 2/2004, July
Recent changes to the AGOA legislation
by Eckart Naumann
TB 3/2004, August
2005
Trade after Preferences: a New Adjustment Partnership?
by Ron Sandrey
TB1/2005, June
TRIPs and Public Health: The Unresolved Debate
by Tenu Avafia
TB2/2005, June
Daring to Dispute: Are there shifting trends in African participation in WTO dispute settlement?
by Calvin Manduna
TB3/2005, June
South Africa’s Countervailing Regulations
by Gustav Brink
TB4/2005, August
Trade and competitiveness in African fish exports: Impacts of WTO and EU negotiations and regulation
by Stefano Ponte, Jesper Raakjær Nielsen, & Liam Campling
TB5/2005, September
Geographical Indications: Implications for Africa
by Catherine Grant
TB6/November
2006
Southern Africa and the European Union: the TDCA and SADC EPA
by Catherine Grant
TB1/2006, May
Safeguarding South Africa’s clothing, textile and footwear industries
by Gustav Brink
TB2/2006, May
Agricultural Safeguards in South Africa
by Gustav Brink
TB3/2006, May
The WTO Trade Policy Review Mechanism: application and benefit to SACU
by Paul Kruger
TB4/2006, June
Amendment to TRIPs agreement: consensus or dissension?
by Madalitso Mutuwazovu Mmeta
TB5/2006, September
2007
Southern Africa and the trading relationship with the European Union
Ron Sandrey and Taku Fundira
TB1/2007, January
The development pillar of the EPA negotiation
Catherine Grant
TB2/2007, February
The use and limitations of computer models in assessing trade policy
Ron Sandrey
TB3/2007, March
Competition and infant industry protection within SACU: the case of UHT milk in Namibia
Omu Kakujaha-Matundu
TB4/2007, March
Sunset Reviews in South Africa: New Direction given by the High Court
Gustav Brink
TB5/2007, July
South African quotas on Chinese clothing and textiles: has there been sufficient economic justification?
Johann van Eeden and Ron Sandrey
TB6/2007, September
Sunset reviews in South Africa: how long is five years?
Gustav Brink
TB7/2007, November
The four pillars of South African agricultural trade policy
Ron Sandrey
TB8/2007, November
Update: South African quotas on Chinese clothing and textiles: has there been sufficient economic justification?
Taku Fundira
TB09/2007, December
Countervailing Reviews: Countering Subsidised Exports or Countering Subsidy Programmes?
Gustav Brink
TB10/2007, December
Books
2007
Monitoring Regional Integration in Southern Africa: Volume 6 (2006)
Editors: Anton Bösl, Willie Breytenbach, Trudi Hartzenberg, Colin McCarthy, Klaus Schade
February 2007, ISBN 978-0-9584680-5-3
Economic Partnership Agreements: Handbook
Authors: Talitha Bertelsman Scott and Catherine Grant
April 2007
South Africa's way ahead: trade policy options
Authors: Ron Sandrey, Hans Grinsted Jensen, Nick Vink, Taku Fundira
September 2007, ISBN 978-0-9584680-7-7